Spillane y Orlina LPS - 05 Investigating Leadership Practice - Distributed Perspective

This article was downloaded by: [Universidad de Sevilla]
On: 19 April 2010

Access details: Access Details: [subscription number 773444416]
Publisher Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
Leadership and Policy in Schools

Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713734379
Investigating Leadership Practice: Exploring the Entailments of Taking a

Distributed Perspective
James P. Spillane a;Enrique C. Orlina a
a
Northwestern University,
To cite this Article Spillane, James P. andOrlina, Enrique C.(2005) 'Investigating Leadership Practice: Exploring the
Entailments of Taking a Distributed Perspective', Leadership and Policy in Schools, 4: 3, 157 — 176
To link to this Article: DOI: 10.1080/15700760500244728
URL: http://dx.doi.org/10.1080/15700760500244728
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Leadership and Policy in Schools, 4:157–176, 2005
Copyright © Taylor & Francis Inc.
ISSN: 1570-0763 print/1744-5043 online
DOI: 10.1080/15700760500244728
Investigating Leadership Practice: Exploring

1744-5043 and Policy in Schools
1570-0763
NLPS
Leadership Schools, Vol. 04, No. 03, August 2005: pp. 0–0
the Entailments of Taking a Distributed

Perspective
JAMES P. SPILLANE AND ENRIQUE C. ORLINA

Investigating
James P. Spillane
Leadership
and Enrique C. Orlina
Northwestern University
This paper examines two questions: What does it mean to take a

Downloaded By: [Universidad de Sevilla] At: 18:37 19 April 2010
distributed perspective on leadership in schools? What are the

entailments of taking a distributed perspective for research on
school leadership? Arguing that the practice of leadership
should be a key concern in scholarship on school leadership,
the authors explore the leader-plus and the practice aspects of
taking a distributed perspective. Arguing that both aspects are
essential when taking a distributed perspective, the authors
review findings from recent research that dwells mostly on the
leader-plus aspect. Turning to the second question, the authors
explore the methodological challenges involved in taking a dis-
tributed perspective.
An earlier version of this paper was presented at the annual meeting of the American
Educational Research Association, Montreal, April 2005. This paper is based on research
supported by the Distributed Leadership Studies (http://www.distributedleadership.org),
with research grants from the National Science Foundation (REC-9873583), the Spencer
Foundation, the Carnegie Foundation, and the Rockefeller Foundation. Northwestern
University’s School of Education and Social Policy and Institute for Policy Research also
supported the work. This paper would not have been possible without the Distributed
Leadership Study’s interdisciplinary research team of postdoctoral fellows, graduate
students, and undergraduate students. All opinions and conclusions expressed in this
paper are those of the author and do not necessarily reflect the views of any funding
agency.
Address correspondence to James Spillane, Distributed Leadership Studies Principal
Investigator, School of Education and Social Policy, Northwestern University, 2020 Campus
Drive, Evanston, IL, 60201. E-mail: j-spillane@northwestern.edu
157
158 James P. Spillane and Enrique C. Orlina
INTRODUCTION
Distributed leadership has garnered the attention of many scholars, practitio-

ners, and policymakers. Its appeal in part lies in the ease with which it can
become many or all things to all people, often used as a synonym for demo-
cratic leadership, participative leadership, collaborative leadership, and so on.
It is no surprise then that some observers wonder if distributed leadership is
just another case of old wine in new bottles. Some talk about distributed lead-
ership as a recipe for effective school leadership, others as a conceptual lens
for framing investigations of leadership practice. Hence, some observers
question whether there is any real substance to a distributed leadership per-
spective. Does the emperor have any new clothes (Spillane, 2005a)
In this article we take up two questions. Our first question is this: What
does it mean to take a distributed perspective on school leadership? With a
quick slight of hand, we answer this question by describing our understanding

of a distributed perspective based on the Distributed Leadership Study
(Spillane, 2005a; Spillane, Halverson, & Diamond, 2001, 2004). Our second
question is this: What are the entailments of taking a distributed perspective
to investigate school leadership?
Our article is organized as follows. Acknowledging the centrality of
practice in efforts to understand leadership from a distributed perspective,
we examine the meaning of the term practice. Next, taking the position that
a distributed perspective is not a blueprint for leadership but rather a frame-
work for thinking about leadership, we identify two essential aspects of this
perspective: the leader-plus aspect and the practice aspect. We argue that
while much of the literature dwells on the leader-plus aspect, both aspects
are essential to our understanding of what it means to take a distributed
perspective. We devote much of our discussion to the practice aspect. We
also explore how a distributed perspective, as we understand it, is not syn-
onymous with democratic leadership and other popular “approaches” to
leadership. Then, turning to our second question, we examine the method-
ological challenges in taking a distributed perspective when studying school
leadership.
LEADERSHIP AND LEADERSHIP PRACTICE
Definitions of leadership are plentiful. Some scholars have labored in the

definitional labyrinth, often waylaid in efforts to define or defunct defini-
tions of leadership. Such endeavors have their place. We take a more prag-
matic approach, adopting a working definition of leadership that has been
adapted and revised based on our empirical work. “Leadership refers to
activities tied to the core work of the organization that are designed by
organizational members to influence the motivation, knowledge, affect, and
Investigating Leadership Practice 159
practices of other organizational members or that are understood by organi-

zational members as intended to influence their motivation, knowledge,
affect, and practices” (Spillane, 2005a, p. 11). Defining leadership in this
way means that all influence relations in schools are not leadership.
Leadership is reserved for those activities that administrators and teach-
ers either design to influence others, or that others understand as intended
to influence them, in the service of the organization’s core work. We
acknowledge here that the “organization’s core work” poses some chal-
lenges. While leadership is frequently designed to initiate change, it can
also be about preserving the status quo in the face of change efforts
(Cuban, 1988).
By defining leadership in this way, we part company with two features
of many other definitions. First, in some definitions the term leadership is
limited to its sense as something that has been accomplished: evidence of
someone having actually been influenced by someone else is necessary to

denote leadership. We find this problematic. Specifically, in the Distributed
Leadership Study we find instances where even when teachers ignore the
guidance offered through leadership practice, they still understand these
practices as leadership—designed to influence what they do. People can
construct activities as leadership even if the influence associated with those
activities does not lead them to change. Our definition of leadership thus
leaves room for qualifiers like “effective” or “ineffective,” words that capture
the extent to which an attempt to influence produces the intended results.
Second, some confine the term leadership to social influence relations that
have positive outcomes. But, leadership needn’t necessarily involve outcomes
or processes that are “positive” or “beneficial.” We believe that questions of
the “effectiveness” and “direction” of influence must be separated from
defining the leadership phenomena itself.
We can distinguish between leadership as practice and leadership as an
enterprise.1 When we say that someone is leading a particular school or that
the leadership at a particular school is exemplary, we are talking about
leadership as an enterprise. On the other hand we might also talk about a
particular instance of leadership, by which we mean leadership practice.
Not everything that school leaders do while in school is classified as leader-
ship practice.
For us leadership practice refers to what is done in a particular time
and place to act in response to what Bourdieu terms “the urgency of prac-
tice” (1981, p. 310). The imperative nature of practice limits reflection and
the taking of time to weigh options, reinforcing the importance of “habitus”
(Bourdieu, 1990). Habitus underscores that individuals develop dispositions
1
We borrow this distinction from Phil Jackson (1986) and Paul Komisar (1968) and their work on
teaching as activity and teaching as enterprise.
to act in one way or another in certain situations and these dispositions may
not be at the level of conscious decisions (Bourdieu, 1990; Foucault, 1977).
Habitus underscores that how we interact in the world is in part a function
of our own past and the pasts of the groups and institutions to which we
belong. Things happen and people act, but only in relation to others, hence
our attention to interactions (see next section). Practice is embedded in
time, “inseparable from temporality,” and therefore cannot be fully under-
stood outside of its relations to time and place (Bourdieu, 1990, p. 81).
Hence, a description of a behavior on its own is insufficient in understand-
ing leadership practice.
Time is measured with widely varying scales, ranging from the
moment-by-moment interactions in a particular place to historical time. Our
attention to practice as it unfolds in the moment-by-moment interactions in
a particular place and time is not meant to privilege the micro at the
expense of the macro. Practice in a particular place and time has to be

understood as part of systems of practice or activity systems (Gronn, 2003;
Engestrom, 1999a).
Bourdieu also stresses the logic of practice, a logic that “is able to orga-
nize all thoughts, perceptions and actions by means of a few generative
principles” (1990, p. 86). People’s logic emerges from a life of negotiating
social contexts and is realized in ongoing interactions through a “regulated
improvisation” (p. 57). This again highlights the importance of time, as
improvization is as much about responding to what came before as it is
about dealing with issues of the immediate moment. It should be empha-
sized that people are not fully constrained by the situation and their inter-
nalized logics. Bourdieu includes in his thinking about practice the idea that
people have goals and interests and thus that they strategize about how to
achieve those goals. People are seen to have a grasp on “the usual pattern
of how things are done or happen” (Jenkins, 2002, p. 72) and to use that
knowledge to guide their behaviors.2 Hence, we believe that the key to
understanding leadership practice is to understand how it arises out of peo-
ple’s ongoing attempts to negotiate their relationship with their situations.
Some scholars lament the inattention to the practice of leadership.
Arguing that an “action perspective sees the reality of management as a
matter of actions,” Eccles and Nohria encourage an approach to studying
leadership that focuses on action rather than exclusively on structures,
states, and designs (1992, p. 13). Putting practice at the forefront is critical
because “the strength of leadership as an influencing relation rests upon its
effectiveness as activity” (Tucker, 1981, p. 25). In education, Heck and
Hallinger (1999) argue that in-depth analysis of leadership practice is rare
2
An in-depth discussion of the various theoretical arguments around practice are beyond the scope
of this paper.
but essential if we are to make progress in understanding school leadership.

For some leadership scholars, such claims may appear somewhat over-
drawn. After all, there is an extensive literature on leadership styles, models,
and approaches built up from analysis of collections of coherent practices.
While such attempts to aggregate behaviors into broad patterns are helpful,
they extract leadership practice from its temporal context and in so doing
ignore “the urgency of practice” (Bourdieu, 1981, p. 310).
The confusion with respect to the attention given to practice results in
part because the term is used in different ways in writings on school leader-
ship, with the intended meaning of the word often left implicit. Practice is
used to refer to the comprehensive enactment of the profession, a set of
specific skills or behaviors, the counterpart of theory, and the actual doing
of leadership in particular places and times.
A central argument threaded throughout this paper is that the practice
of leadership in a particular place and extending over time is a critical line

of inquiry for scholars. Knowing what leaders do is one thing, but a rich
understanding of how, why, and when they do it is essential if research is to
make a meaningful contribution to understanding and improving leadership
practice. While an extensive literature on leadership antecedents attempts to
account for leaders’ practice in terms of their cognitions, problem solving, emo-
tions, and other “traits,” we argue that these variables are insufficient on their
own if we take seriously that practice unfolds over time in the interactions
among leaders and their situation, nested in systems of practice. An in-
depth analysis of the practice of leadership, not just the practice of leaders,
is an area meriting attention of scholars of school leadership in particular
and organizational leadership writ large.
To illuminate what we mean by practice, consider building a culture
that supports collaboration among teachers, an organizational function
thought critical for school improvement (Firestone, 1989). Recent scholar-
ship suggests that school leaders cultivate these conditions by creating con-
versation networks that support ongoing deliberations about practice (Louis
et al., 1996). By setting tasks that involve teachers working together, princi-
pals can support norms of collaboration (Goldring & Rallis, 1993). While
this work makes a valuable contribution to scholarship on school leader-
ship, in our scheme it is not an analysis of leadership practice but rather of
leadership functions and strategies. A focus on practice highlights the imme-
diacy of the interactions to which organizational actors must be sensitive
and responsive. When a function like building a collaborative culture is
reduced to a set of programmable strategies, abstracted out of the dynamic
situations in which culture and practice actually operate, it becomes discon-
nected from the urgency of leadership practice. Our scheme embraces that
urgency in an effort to understand how these functions and strategies play
out in day-to-day practice—how leadership practice unfolds in time to cre-
ate and nurture tasks that get teachers working together.
ELEMENTS OF A DISTRIBUTED PERSPECTIVE3
From a distributed perspective, leadership practice takes place in the

interactions of people and their situations. Leadership practice cannot be
understood by focusing exclusively on the actions of individual leaders—
the interactions out of which leadership practice arises are critical. Although
the idea of leadership practices as a set of skills and behaviors is valuable,
we believe that an important shortcoming of the skills-and-behaviors
approach in efforts to understand practice is that it decontextualizes school
leadership, masking critical interdependencies among people and aspects of
their situation.
A distributed perspective on leadership involves a leader-plus aspect
and a practice aspect (Spillane, 2005a). The leader-plus aspect is essential
but insufficient. The leadership practice aspect foregrounds the interactions
among leaders and followers, and their situation. The practice aspect moves
us beyond simply aggregating the actions of individual leaders.
The Leader-Plus Aspect

Both formally designated leaders and informal leaders take responsibility for
leadership. The leader-plus aspect acknowledges the work of all individuals
who have a hand in leadership practice, whether or not they are formally
designated as leaders.
In addition to school principals and assistant principals, other formally
designated leaders and teachers take responsibility for leadership routines and
functions (Camburn, Rowan, & Taylor, 2003; Firestone, 1989; Hargreaves &
Fink, 2004; Heller & Firestone, 1995; Portin et al. 2003). The distribution of
responsibility for school leadership differs depending on the particular lead-
ership function or routine (Camburn et al., 2003; Heller & Firestone, 1995;
Spillane, Diamond et al., 2004). School principals perform a broader range
of functions, including instructional leadership functions, building manage-
ment functions, and boundary-spanning functions, compared with occu-
pants of other formally designated leadership positions (Camburn et al.,
2003).
Depending on the subject matter, the distribution of leadership among
formally designated leaders and teachers also differs (Spillane, 2005b). For
example, in elementary schools the performance of leadership routines
such as professional development for language arts typically involves more
leaders—both formally designated and teacher leaders—than similar leader-
ship routines related to mathematics (Spillane, 2005b). The distribution of
responsibility for leadership among formally designated leaders and teachers also
3
This section draws from Spillane, 2005a.
differs by the type of school (Portin et al., 2003), school size (Camburn et al.,
2003), and the school or leadership teams developmental stage (Copland,
2004; Harris, 2002).
Responsibility for leadership can be distributed in at least three
arrangements—strict division of labor, co-performance, and parallel perfor-
mance—that can coexist in the same school (Spillane, 2005a). For the most
part, a single leadership position rarely takes responsibility for a particular
leadership function, so a neat division of labor for leadership work is the
exception rather than the norm in schools (Heller & Firestone, 1995). The
co-performance arrangement involves two or more leaders performing a
leadership function or routine together and in a collaborated fashion (Spillane,
Diamond, & Jita, 2003). Leaders don’t always work in a collaborated manner;
they often work in parallel to execute the same leadership functions or rou-
tines, duplicating each other’s work (Heller & Firestone, 1995).
Each of the three arrangements outlined above can involve leaders

striving for similar goals. However, the arrangements can also involve lead-
ers striving for different or even conflicting goals as leaders take responsibility
for the same leadership routine (Spillane, 2005a). Leaders can and do pull in
different or even opposing directions while working together or in parallel.
A leader or group of leaders in a school can push one instructional vision
while another leader or group of leaders attempts to sell an altogether dif-
ferent vision. For example, the new principal and assistant principal of a
school in the Distributed Leadership Study worked together to institute new
routines for classroom observation while teacher leaders were working in
parallel to reinforce the existing norm of teacher autonomy and freedom
from intrusion. From a distributed perspective, each of these activities,
occurring in parallel, is a distinct instance of leadership practice. In contrast
to this scenario, leaders can collaborate in the co-performance of one lead-
ership routine and still promote different goals, such as different visions for
mathematics and literacy instruction. In sum, individuals can perform the
same leadership routine or work on the same leadership function, but do so
in ways that are intended to promote different or even contrary goals.
Similarly, the strategies that leaders use or the roles they take on as
they co-perform or work in parallel in the execution of a leadership routine
can also differ from one leader to the next. Different means, however, are
not necessarily contrary. Take the good cop, bad cop situation. Both cops
have can have the same goal—to get the prisoner to confess—but adopt
very different roles involving different strategies as they co-perform the
interrogation. Similar situations are possible when two or more individuals
work on the performance of the same leadership routine.
Research further suggests that the distribution of responsibility for lead-
ership can come to pass through four mechanisms—by design, by default,
by evolution, and by crisis—that are not mutually exclusive (Spillane,
2005a). The deliberate design decisions of formal and informal leaders
contribute to the distribution of responsibility for leadership work among

formally designated and informal leaders. Alternatively, the distribution of
responsibility for leadership functions and routines can arise by default
when formal leaders or teacher leaders, either individually or collectively,
take on responsibility for some leadership routine or function (Spillane,
2005a). Subsequently, the distribution of leadership among leaders evolves
over time as individuals get to know one another’s skills and weaknesses,
develop trust, and create working relations that contribute to the distribu-
tion of leadership (Gronn, 2003). Finally, the distribution of leadership can
emerge by crisis when a school encounters an unanticipated problem or
challenge, and formal leaders and teachers find themselves working
together to address it (Gronn, 2003).
Reading much of the literature on distributed leadership, one is often left
with the impression that the leader-plus aspect is the extent of what is involved
in taking a distributed perspective. However, the leader-plus aspect is only part

of what it means to take a distributed perspective. A distributed perspective
goes beyond simply enumerating multiple leaders and documenting their con-
tributions; it is foremost about leadership practice (Spillane, 2005a).
The Practice Aspect

From a distributed perspective, leadership practice is central; the perspec-
tive focuses attention on the practice of leadership and not just who takes
responsibility for leadership functions and routines. Most scholarship
equates leadership practice with the actions of individual leaders. In con-
trast, from a distributed perspective, leadership practice is framed in a
particular way: as a product of the interactions of school leaders, follow-
ers, and aspects of their situation (see Figure 1) (Gronn, 2000; Spillane,
Halverson, & Diamond, 2001). The additive models to which the leader-
plus aspect lends itself are insufficient in that they fail to notice and inves-
tigate the interactions among leaders, and among leaders, followers, and
their situation, that give form to leadership practice. In Figure 1 leadership
practice is represented by a triangle, with each angle representing one of
the three essential elements (Spillane, 2005a). A single triangle is used to
represent the interactions among leaders, followers, and situation at a par-
ticular moment in time. The multiple triangles represent the multiple inter-
actions involved in the performance of a leadership routine over time.
Again, time is critical. The broken lines connecting the triangles denote
that the connections between these interactions can be more or less tight
over time.
The empirical knowledge base on the practice aspect of distributed
leadership is thin because few scholars have investigated how leadership
takes shape in the interactions among leaders, followers, and their situation.
FIGURE 1 Leadership Practice (taken from Spillane, 2005).
We briefly review what the existing work tells us by first looking at the peo-
ple dimensions and then examining the situation dimension.
People in Interaction
From a distributed perspective, a critical challenge involves unpacking how
leadership practice is stretched over the interaction among leaders. One
way to do this is by analyzing the interdependencies among leaders’ actions
(Gronn, 2003; Spillane et al., 2003).
By applying a distributed perspective to analyze the interactions among
leaders in the co-performance of leadership practice, researchers in the Dis-
tributed Leadership Study have identified three types of distribution:4
• Collaborated distribution refers to leadership practice that is stretched

over the work of two or more leaders who work together in place and
time to perform the same leadership routine, such as conducting a profes-
sional development workshop. In this situation the practice is similar to
basketball, where players interact with each other as they pass the ball to
teammates when they stop dribbling and work to set one another up to
score.
4
See Spillane, Diamond, & Jita, 2003; Spillane, Diamond, Sherer, & Coldren, 2004.
• Collective distribution refers to leadership practice that is stretched over

the work of two or more leaders who perform a leadership routine while
working separately but interdependently. The interdependencies here are
more like baseball or cricket, where players at bat perform alone, but
their actions in interaction with that of the pitcher or bowler collectively
produce the practice.
• Coordinated distribution refers to those leadership routines that have two
or more activities that have to be performed in a particular sequence. In
this situation, the interdependency is similar to a relay race, in that the co-
performance of the relay race is dependent on a particular ordered
sequence.
These different types of distribution are not mutually exclusive; a single

leadership routine could involve more than one type. Peter Gronn offers a
different typology involving co-performance and collective performance,

which are roughly analogous to what I term collaborated and collective
distribution, respectively (Gronn, 2003). Gronn’s typology also focuses both
on the types of distribution and their origins.
As we argued earlier, leaders do pull in different and sometimes
opposing directions even when they work on the performance of the same
leadership routine. This happens when leaders are working in parallel and
when they co-perform leadership routines. One group of leaders in a school
can promote one vision for classroom instruction and the way it should be
managed, while another group of leaders sell an altogether different vision
and set of arrangements for managing instruction. A distributed perspective
can be applied in analyzing practice in situations where leaders are pulling
together and situations where they are pulling in different or even opposing
directions. From a distributed perspective, the challenge is to understand
how practice takes shape in the interactions of leaders, followers, and their
situation, recognizing that whether they seek similar, different, or even con-
flicting goals is but one dimension of any such analysis.
These types of distribution are not necessarily approaches to leadership
that one “activates” by “doing distributed leadership.” Instead, these are
ways of characterizing the practice that takes shape as school leaders inter-
act in the performance of leadership routines. Each type of distribution
involves different sorts of interdependencies that pose particular challenges
for leadership practice. For example, situations involving collaborated distri-
bution require leaders to co-perform in public, accentuating the affective
dimension of interactions among leaders, whereas situations involving col-
lective distribution allow leaders to co-perform separately, potentially
downplaying the affective dynamic. Hence, situations involving collabo-
rated distribution may necessitate more attention to the affective dimension.
It is imperative to consider the collective rather than relying entirely on
characterizing the interdependencies among individuals. If an important
unit of analysis for examining leadership practice is leaders working with

followers mediated by aspects of the situation, then analyzing the interde-
pendencies only gets us so far. In analyzing interactions among leaders in
the co-performance of leadership practice, it is necessary to attend to both
the workings of the group or collective and its component parts. A critical
issue concerns how leaders (and indeed followers) manage to work
together.
Karl Weick and his colleagues construct of heedfulness offers some
theoretical leverage here (Weick & Roberts, 1993; Weick & Sutcliffe, 2001).
Individuals who act like a group “interrelate their actions with more or less
care,” and it is in this interrelating that we can begin to identify how
the group functions as a collective (Weick & Roberts, 1993, p. 360).
Heed describes the way in which a set of behaviors is performed:
groups act heedfully when they act carefully, intelligently, purposefully, and
attentively.
Group members engaged in heedful interrelations have a sense of
themselves as an ensemble. There are three interrelated processes involved
in groups acting more or less heedfully. To begin with, group members cre-
ate the social norms between group members by acting as though these
norms exist. Acting as though there are social norms, people construct their
actions by envisioning a system of joint action and connect that constructed
action with the system they envisioned. The result is a system of practice
that does not reside entirely in the individuals but also in the interrelating or
interacting between them. The last point, in the context of our current dis-
cussion, reinforces the fact that leadership is stretched over leaders and not
just a sum of individual contributions. We prefer the terms “group practice”
or “co-performance” to the term “joint action.” With respect to the interac-
tions among school leaders in a faculty meeting, for example, we might
argue that the participants construct a taken-as-shared field within which
members of the group improvise, playing off each other’s actions. We
should point out that in our understanding, leaders can have a sense of
themselves as a group and engage in heedful interactions with one another
in the performance of a leadership routine and still strive for different or
even contrary outcomes through that routine.
The three types of distribution pose different challenges with respect to
the heedfulness of the interrelating among participants. In collaborated dis-
tribution situations, for example, leaders co-performing a leadership routine
have multiple opportunities to assess their colleagues’ perspectives and
actions firsthand. These circumstances can increase the opportunities for
more heedful interrelating among leaders. In contrast, situations involving
collective distribution, where leaders may have fewer opportunities to
observe each other in action, offer fewer opportunities for increasing the
heedfulness of the interacting. Indeed, the likelihood of heedless interacting
is perhaps greater in situations involving collective distribution compared
with those involving collaborated distributions. The heedfulness of the

interacting among leaders in collective distribution situations depends on
the accuracy of leaders’ reports to each other about what they do separately
but interdependently.
Of course, followers are also critical in understanding how leadership
takes shape in interaction among people. While some scholars question the
leader-follower distinction, our fear is that if the distinction is ignored
altogether the role of followers in defining leadership practice will be insuf-
ficiently understood. The nature of the interactions among leaders
and among leaders and followers differs depending on the subject area
(Spillane, 2005b), another finding that follows from taking a distributed per-
spective with an explicit accounting of the role of the follower.
People and Place

Scholars have long recognized that the situation is critical to all sorts of
practice, including leadership practice. Social scientists have a fondness for
pointing out the importance of situation, though what counts as situation is
more often than not broadly defined. Most of us are well aware of how the
situation impinges on our work and play. Situation can make practice more or
less difficult. We more often than not think of the situation as an accessory
to practice—the tool allows us to perform more or less efficiently or effec-
tively. Aspects of the situation, however, do more than quicken and smarten
up our execution or performance of some practice.
Tools, routines, and other aspects of the situation can be viewed as
mediating human action (Wertsch, 1991). These factors thus serve as essen-
tial go-betweens in our interactions with others, and, as argued previously,
it is in such interactions that practice takes shape. Therefore, situational fac-
tors do not simply influence practice from the outside in by enabling people
to execute their plans more or less efficiently. Instead, they they contribute
to defining and giving form to leadership practice by their mediating role in
leader-follower interactions. Tools and routines and other aspects of the sit-
uation put some elements of interactions at the forefront while background-
ing or obscuring others. They thereby help define what leaders and
followers must be heedful of in their interactions.
By way of example, consider two groups of teachers who meet reg-
ularly to as part of their respective school’s efforts to develop teachers’
skills in mathematics teaching. Both groups are facilitated by a mathe-
matics teacher leader and a mathematics specialist and, for argument
sake, let’s assume that the groups and mathematics curricula are identi-
cal in all respects with one exception. Group A uses samples of chil-
dren’s mathematics work in their discussions along with the students’
test scores. Group B uses only student test scores. We argue that the
tools in play in Group A—student work and student test scores—is likely
to frame the interactions among leaders and followers around not only
which children do well on which topics but also on the children’s math-
ematical thinking and strategies that might account for their perfor-
mance. In contrast, in Group B discussions are more likely to focus more
on which children do well on which topics. The tools used in leadership
practice enable and constrain the interactions among leaders and follow-
ers. In this way, different tools contribute to defining leadership practice
differently. Hence, new or redesigned tools may change practice, just as
we hope that new knowledge and skill on the part of leaders might
change that practice. However, we emphasize that the tools do not
determine practice. The situation is also a product of the interactions
among leaders and followers. It is both the medium for and the outcome
of leadership practice. From a distributed perspective, the challenge
involves describing how aspects of the situation such as tools enable and
constrain leadership practice.

The practice aspect of a distributed framework captures the importance
of going beyond an accounting of individual leaders and the actions they
take. Further, simply acknowledging that the distributed leadership perspec-
tive emphasizes leadership practice does not sufficiently differentiate the
distributed perspective from other approaches to describing or studying
leadership.
METHODOLOGICAL CONSIDERATIONS
Adopting a distributed perspective on leadership poses both epistemo-

logical and methodological challenges, especially if we take seriously
the practice aspect of the distributed framework. In this section we con-
sider the challenges involved in identifying and documenting leadership
practice.
Identifying Leadership Practice

Even if one can operationalize a definition of leadership practice, using a
distributed perspective to investigate school leadership poses a number of
methodological challenges. One of these challenges concerns the identifica-
tion of leadership in schools. If we take a distributed perspective, then rely-
ing on a definition of leadership practice as the practice or practices of one
or more formally designated leaders is insufficient. Often, investigators rely
on labor intensive and costly ethnographic and structured observation
methods to identify leadership practice. While these methods generate rich
insights into leadership practice, they are typically limited by small sample
sizes that make it difficult to generalize to a population of schools. Hence,
an exclusive reliance on ethnographic methods only gets us so far.
Developing valid and reliable ways to identify leadership practice is

difficult because this practice potentially spans the entire school, involving
the work of not only formally designated leaders but also of individuals
with no such designations. Further, leadership practice is not ordinarily con-
fined to structured and patterned time slots; it happens before and after
school, on the weekends, and during school hours in both formally desig-
nated routines, such as grade-level meetings, and informal interactions, such
as over lunch or in exchanges snatched between classes. While challenging,
the task of developing more robust means of identifying leadership practice
is not insurmountable.
We approach the task of identifying leadership practice from two
angles simultaneously—from the perspective of those in formally desig-
nated leadership positions and from the perspective of followers. We are
developing a leadership inventory designed to get formally designated
leaders in a school to identify the formal and informal routines they per-
form that are intended to influence the motivation, knowledge, affect, and
practices of other organizational members. At the same time, to get at
leadership practice from the followers’ perspective (we assume everyone
in the organization is potentially a follower at some point, including for-
mally designated leaders), we are using a modified social network survey
instrument.
Our goal with the social network instrument is not only to get at
activities that are intended to influence other organizational members, but
also to get at those activities that are understood by organizational mem-
bers as intended to influence their motivation, knowledge, affect, and
practice. The social network instrument explores multiple dimensions of
leadership for mathematics instruction, including a) who respondents go
to for advice/knowledge about different dimensions of instruction (e.g.,
content, pedagogy, materials), and b) who respondents trust with an
instructionally sensitive problem (e.g., not understanding the mathematics
in the curriculum). While social network instruments rely on the identifi-
cation of people, our instrument is designed to also identify leadership
routines (e.g., grade-level meetings) and artifacts (e.g., curriculum guides)
that respondents perceive as either intending to and/or actually influenc-
ing their motivation, knowledge, affect, and practice. Our initial piloting
of the social network survey at the elementary level identified substantial
differences with respect to leadership for mathematics and language arts
(Spillane, 2005b).
Documenting and Analyzing Leadership Practice

To date, scholars and practitioners have relied mostly on either ethno-
graphic and structured observational methods (e.g., shadowing) or surveys
to document and analyze leadership practice. Both have distinctly different
problems. The ethnographic approach and shadowing typically are too

expensive to use in more than a handful of schools. The survey approach
fails to capture the nuances of leadership practice, especially the differences
between reported practice and actual practice, and fails to get at interac-
tions. A key challenge for scholars of leadership practice will involve devel-
oping instruments that allow us to capture leadership practice across large
samples of schools. We are currently developing and validating a series of
logs—daily practice logs, experienced sampling method logs, and event
logs—to document leadership practice.
Still, if interactions are indeed a critical unit of analysis, then the chal-
lenge for scholars of leadership practice involves documenting and analyz-
ing interacts rather than just acts. It is not sufficient to simply log formally
designated and informal leaders to identify which leaders interact in the
performance of which leadership routines while using which tools. More is
necessary. Weick (1979) describes the interact as an essential element in

social organizational processes. He further notes that the appropriate unit
of analysis for studying organizations is the double interact, “the basic unit
for describing interpersonal influence” (p. 89). He defines these concepts as
follows:
[Social] processes contain individual behaviors that are interlocked

between two or more people. The behaviors of one person are contin-
gent on the behaviors of another person(s), and these contingencies are
called interacts. The unit of analysis in organizing is contingent
response patterns, patterns in which an action by actor A evokes a spe-
cific response in actor B (so far this is an interact), which is then
responded to by actor A (this complete sequence is a double interact).
(Weick, 1979, p. 89, italics in original)
Thus, Weick proposes that interacts and double interacts are the key to
understanding organizational processes.
We believe that interacts and double interacts are also an essential
component of investigating leadership from a distributed perspective. In
particular, as noted above, double interacts are the basic unit for investigat-
ing interpersonal influence, a fact that is extremely important given our def-
inition of leadership as interpersonal influence in service of the core task of
the organization. Double interacts capture influence in terms of action/
response/secondary response sequences. For example, consider an initiat-
ing act in which person A proposes a new interpretation of an event that
occurred in the school. The response might then be for person B to propose
an alternative interpretation. This, in Weick’s terms, is an interact. If we then
consider person A’s secondary response—for example, person A might
change positions and adopt person B’s interpretation—then we have
defined a double interact, one in which person B has influenced person A’s
position. We believe that by extending this kind of analysis to school lead-

ership teams in context, we can begin to better understand the practice of
school leadership.
Analytically, the measurement and analysis of interacts poses some sig-
nificant challenges. In fact, we have found few examples where this
approach has been applied to the study of group phenomena in real, as dis-
tinct from experimental, situations. One notable series of studies investi-
gated the way that groups made decisions (Poole, 1981, 1983a, 1983b;
Poole & Roth, 1989a, 1989b), finding that contrary to prior models in which
groups followed a rigid and predictable sequence of phases in arriving at a
decision, the process was instead divided into multiple smaller phases that
might occur in any order for any given group. Central to the analysis of
group decisions was the coding and analysis of transcribed group meetings
in terms of interacts. The analysis enabled Poole and Roth to identify phases
of group-decision tasks according to the patterns of interacts that dominated

the group processes.
Since Poole published his original study, additional analytical techniques
have been introduced. Bakeman and colleagues (Bakeman & Gottman,
1997; Bakeman & Quera, 1995) developed a method of sequential analysis,
a statistical approach for examining data that preserves the temporal order
(and in some cases, the precise durations) of events in social interaction.
Using sequential analysis, one can determine the frequency with which a
given act occurs, given that an initiating act has occurred. The situation just
described is a lag-1 analysis: the frequency of a target act one time slot after
a given act. The analysis can be extended to look at longer lags (e.g., the
frequency of the target act two time slots after the given act) or even nega-
tive lags (i.e., the frequency that the target act occurred immediately prior to
the given act). Again, we believe that by applying techniques such as this to
data from school teams in interaction (we are analyzing data from the Dis-
tributed Leadership Study), we can better understand the nature of leader-
ship practice as it occurs in interactions. Our strategy here is to sample from
the population of particular leadership routines (e.g., grade-level meetings)
over the course of a school year and analyze the interactions.
The analysis approach we are pursuing is in direct response to the dis-
tributed leadership perspective that frames our work and reflects our con-
ceptualization of practice described above. Practice occurs in time; the
duration, pacing, and sequence of events matters. Moreover, it is the pattern
of behaviors—the ways in which people react to and are guided by their
environments—that is revealing of the underlying logic of practice.
We find at least two shortcomings with Weick’s work and work on analyz-
ing interacts and double interacts in general. First, work on coding interactions
pays scant attention to artifacts and tools of various sorts. However, the distrib-
uted leadership framework identifies artifacts, tools, and other aspects of the sit-
uation as constituting elements of leadership practice. Aspects of the situation
such as tools and routines are the vehicles through which leaders interact with
one another and with followers. For example, tools and routines embody
scripts that capture repeated patterns for these interactions and in actual
practice a routine or the use of a particular tool can be modified and revised.
Hence, in our efforts examine leadership practice through an analysis of the
interactions, we also code for tools, artifacts and other aspects of the situation.
A second shortcoming is that in focusing on interacts one can easily get
lost in the here and now—the micro-analysis of interactions—and forget
that these interacts have to be situated in time and in place. By situating
interactions in time we mean not only the twenty or thirty minutes of a par-
ticular leadership routine, but also situating the interactions in a particular
week, month, semester, school year, and reform period. Moreover, a partic-
ular leadership practice has to be situated in a particular school and in the
broader institutional context of a school district, state, and country. Hence,
while our attention to practice centers on the interactions among leaders,

followers, and their situation in particular places and time, it simultaneously
attempts to examine how social structures at various levels of the system
enable and constrain these interactions. One way in which we do this is by
analyzing how tools and artifacts of various sorts (e.g., district test data,
state standards) contribute to defining leadership practice by framing the
interactions among leaders and among leaders and followers. At the same
time, by examining how artifacts and tools get used over time, we can begin
to unpack how tools and artifacts get redefined in and through practice. In
this way, we strive to analyze the interactions that are the building blocks of
leadership practice while simultaneously attending to the broader institu-
tional context in which this practice is nested.
CONCLUSION
A distributed perspective on leadership is a framework for thinking about

and analyzing leadership. It is not, in itself, a prescription or recipe for how
to lead. It is a conceptual or diagnostic tool for thinking about school lead-
ership in new ways. It offers a conceptual frame for researchers to focus
their investigations and for practitioners to focus their diagnosis of leader-
ship. In our usage, then, a distributed perspective is normative only in the
sense that any theoretical framework is normative in that it foregrounds
some aspects of a phenomenon and backgrounds others.
Distributed leadership is often used interchangeably with collaborative
leadership, democratic leadership, transformational leadership, and situa-
tional leadership, among others. For us, collaborative, democratic, and other
forms of leadership are not synonyms for distributed leadership. A distrib-
uted perspective allows for leadership that can be more or less collaborative
and that can be either democratic or autocratic. Further, a distributed
perspective on leadership does not privilege a transformational perspective

over a transactional one. Distributed leadership is at best a relative of these
“approaches” to leadership.
Being catholic in our tastes, we are not arguing that a distributed per-
spective is the only way of framing studies of leadership practice. Rather,
we believe that scholars of leadership practice need to articulate clear con-
ceptual frames for studying that practice if the field is to accumulate system-
atic knowledge. Hence, our sketch of what we mean by a distributed
perspective and our discussion of its entailments for research on leadership.
If you take a distributed perspective, the practice of leadership has to
be the central concern. Leadership roles, functions, and structures are
important, but leadership practice is paramount. A distributed perspective
offers a very particular way of thinking about that practice. While school
leaders are important, they are only one of the elements that contribute to
defining leadership practice. Interactions are the key to unlocking leader-

ship practice from a distributed perspective. Leadership practice takes shape
in the interactions of leaders, followers, and their situation. Hence, we have
to start with the leadership at the group level, observe it, infer who the lead-
ers are, and begin to explore the interactions among leaders, followers, and
their situation.
REFERENCES
Bakeman, R., & Gottman, J. M. (1997). Observing interaction: An introduction to

sequential analysis (2nd ed.). New York: Cambridge University Press.
Bakeman, R., & Quera, V. (1995). Analyzing interactions: Sequential analysis with
SDIS and GSEQ. New York: Cambridge University Press.
Bourdieu, P. (1981). Men and machines. In K., Knorr-Cetina, & A. V., Cicourel (Eds.),
Advances in social theory and methodology (pp. 304–317). London: Routledge.
Bourdieu, P. (1990). The logic of practice. Stanford, CA: Stanford University Press.
(R. Nice Trans; original work published in 1980).
Camburn, E., Rowan, B., & Taylor, J. (2003). Distributed leadership in schools: The
case of elementary schools adopting comprehensive school reform models.
Educational Evaluation and Policy Analysis, 25(4), 347–373.
Copland, M. A. (2004). Leadership of inquiry: Building and sustaining capacity
for school improvement. Educational Evaluation and Policy Analysi, 25(4),
375–395.
Cuban, L. (1988). The managerial imperative and the practice of leadership in
schools. Albany, NY: State University of New York Press.
Eccles, R. G., & Nohria, N. (1992). Beyond the hype: Rediscovering the essence of
management. Boston: Harvard Business School Press.
Engeström, Y. (1999a). Activity theory and individual and social transformation. In
Y., Engeström, R., Meittinen, & R. L., Punamäki (Eds.), Perspectives on Activity
Theory. Cambridge: Cambridge University Press (pp. 19–38).
Firestone, W. A. (1989). Using reform: Conceptualizing district initiative. Educa-

tional Evaluation and Policy Analysis, 11(2), 151–165.
Foucault, M. (1977). Discipline and punish: The birth of the prison. London: Allen
Lane. (A. Sheridan, Trans.).
Goldring, E. B., & Rallis, S. F. (1993). Principals of dynamic schools: Taking charge
of change. Newbury Park, CA: Corwin.
Gronn, P. (2000). Distributed properties: A new architecture for leadership. Educa-
tional Management & Administration, 28(3), 317–38.
Gronn, P. (2003). The new work of educational leaders: Changing leadership prac-
tice in an era of school reform. London: Paul Chapman Publishing.
Hargreaves, A, & Fink, D. (2004). The seven principles of sustainable leadership.
Educational Leadership, 61(7), 8–13.
Harris, A. (2002). Effective leadership in schools facing challenging contexts. School
Leadership and Management, 22(1), 15–26.
Heck, R., & Hallinger, P. (1999). Next generation methods for the study of leadership
and school improvement. In J., Murphy, & K., Seashore-Louis (Eds.), Handbook of
research on educational administration (pp. 141–162). San Francisco: Jossey-Bass.

Heller, M. F., & Firestone, W. A. (1995). Who’s in charge here? Sources of leader-
ship for change in eight schools. Elementary School Journal, 96(1), 65–86.
Jackson, P. (1986). The practice of teaching. New York: Teachers College Press.
Jenkins, R. (2002). Pierre Bourdieu (revised ed.). London: Routledge.
Komisar, B. P. (1968). Teaching: Act and enterprise. In C. J. B., Macmillan, &
T., Nelson (Eds.), Concepts of teaching (pp. 63–88). Chicago: Rand McNally.
Louis, K. S., Marks, H., & Kruse, S. (1996). Teacher’s professional community in
restructuring schools. American Educational Research Journal, 33(4), 757–798.
Poole, M. S. (1981). Decision development in small groups I: A comparison of two
models. Communication Monographs, 48, 1–24.
Poole, M. S. (1983a). Decision development in small groups II: A multiple sequence
model of group decision development. Communication Monographs, 50, 206–232.
Poole, M. S. (1983b). Decision development in small groups III: A multiple
sequence model of group decision development. Communication Mono-
graphs, 50, 321–341.
Poole, M. S., & Roth, J. (1989a). Decision development in small groups IV: A typol-
ogy of decision paths. Human Communication Research, 15(3), 323–356.
Poole, M. S., & Roth, J. (1989b). Decision development in small groups V: Test of a
contingency model. Human Communication Research, 15(4), 549–589.
Portin, B., Schneidre, P., DeArmond, M., & Gundlach, L. (2003). Making sense of
leading schools: A study of the school principalship. Seattle, WA: Center for
Reinventing Public Education, University of Washington.
Spillane, J. P. (2005a). Distributed leadership. San Francisco: Jossey-Bass.
Spillane, J. P. (2005b). Primary school leadership practice: How the subject matters.
School Leadership and Management, 25(4), 383–397.
Spillane, J. P., Diamond, J. B., & Jita, L. (2003). Leading instruction: The distribution
of leadership for instruction. Journal of Curriculum Studies, 35(5), 533–543.
Spillane, J. P., Diamond, J., Sherer, J., & Coldren, A. (2004). Distributing leadership.
In M., Coles, & G., Southworth (Eds.), Developing leadership: Creating the
schools of tomorrow (pp. 37–49). New York: Open University Press.
Spillane, J. P., Halverson, R., & Diamond, J. B. (2001). Investigating school leader-
ship practice: A distributed perspective. Educational Researcher, 30(3), 23–28.
Spillane, J. P., Halverson, R., & Diamond, J. B. (2004). Towards a theory of leadership
practice: A distributed perspective. Journal of Curriculum Studies, 36(1), 3–34.
Tucker, R. C. (1981). Politics as leadership. Columbia, MO: University of Missouri
Press.
Weick, K. E. (1979). The social psychology of organizing (2nd ed.). New York:
McGraw-Hill.
Weick, K. E., & Roberts, K. H. (1993). Collective mind in organizations: Heedful
interrelating on flight decks. Administrative Science Quarterly, 38(3), 357–381.
Weick, K. E., & Sutcliffe, K. M. (2001). Managing the unexpected: Assuring high per-
formance in an age of complexity. San Francisco: Jossey-Bass.
Wertsch, J. V. (1991). Voices of the mind: A sociocultural approach to mediated
action. Cambridge, MA: Harvard University Press.

Spillane y Orlina LPS - 05 Investigating Leadership Practice - Distributed Perspective

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spillane y Orlina LPS - 05 Investigating Leadership Practice - Distributed Perspective

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [Universidad de Sevilla]

On: 19 April 2010

Leadership and Policy in Schools

Investigating Leadership Practice: Exploring the Entailments of Taking a

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

Investigating Leadership Practice: Exploring

the Entailments of Taking a Distributed

JAMES P. SPILLANE AND ENRIQUE C. ORLINA

This paper examines two questions: What does it mean to take a

distributed perspective on leadership in schools? What are the

Distributed leadership has garnered the attention of many scholars, practitio-

quick slight of hand, we answer this question by describing our understanding

LEADERSHIP AND LEADERSHIP PRACTICE

Definitions of leadership are plentiful. Some scholars have labored in the

practices of other organizational members or that are understood by organi-

someone having actually been influenced by someone else is necessary to

expense of the macro. Practice in a particular place and time has to be

but essential if we are to make progress in understanding school leadership.

of leadership in a particular place and extending over time is a critical line

ELEMENTS OF A DISTRIBUTED PERSPECTIVE3

From a distributed perspective, leadership practice takes place in the

The Leader-Plus Aspect

Each of the three arrangements outlined above can involve leaders

contribute to the distribution of responsibility for leadership work among

in taking a distributed perspective. However, the leader-plus aspect is only part

The Practice Aspect

FIGURE 1 Leadership Practice (taken from Spillane, 2005).

• Collaborated distribution refers to leadership practice that is stretched

• Collective distribution refers to leadership practice that is stretched over

These different types of distribution are not mutually exclusive; a single

different typology involving co-performance and collective performance,

unit of analysis for examining leadership practice is leaders working with

with those involving collaborated distributions. The heedfulness of the

People and Place

constrain leadership practice.

Adopting a distributed perspective on leadership poses both epistemo-

Identifying Leadership Practice

Developing valid and reliable ways to identify leadership practice is

Documenting and Analyzing Leadership Practice

problems. The ethnographic approach and shadowing typically are too

necessary. Weick (1979) describes the interact as an essential element in

[Social] processes contain individual behaviors that are interlocked

position. We believe that by extending this kind of analysis to school lead-

of group-decision tasks according to the patterns of interacts that dominated

while our attention to practice centers on the interactions among leaders,

A distributed perspective on leadership is a framework for thinking about

perspective on leadership does not privilege a transformational perspective

defining leadership practice. Interactions are the key to unlocking leader-

Bakeman, R., & Gottman, J. M. (1997). Observing interaction: An introduction to

Firestone, W. A. (1989). Using reform: Conceptualizing district initiative. Educa-

research on educational administration (pp. 141–162). San Francisco: Jossey-Bass.

You might also like