MOD5 DISC1

The teacher leadership process: Attempting change within embedded systems

Kristy S. Cooper1 • Randi N. Stanulis1

•

Susan K. Brondyk2 • Erica R. Hamilton3

•

Michael Macaluso1 • Jessica A. Meier1

Published online: 18 November 2015

� Springer Science+Business Media Dordrecht 2015

Abstract This embedded case study examines the leadership practices of eleven

teacher leaders in three urban schools to identify how these teacher leaders attempt

to change the teaching practice of their colleagues while working as professional

learning community leaders and as mentors for new teachers. Using a theoretical

framework integrating complex systems theory with Kotter’s (Leading change.

Harvard University Press, Cambridge, 1996) eight steps for leading organizational

change, we analyze the work and perspectives of individual teacher leaders, and we

examine how teams of teacher leaders and principals function collectively in their

efforts to lead instructional change. Our findings have implications for schools

seeking to utilize teacher leadership as a reform strategy for authentic instructional

improvement.

Keywords Complex systems theory � Instructional improvement � Organizational

change � Professional learning communities � Teacher leadership

Abbreviations PD Professional development

PLC Professional learning community

& Kristy S. Cooper

[email protected]

1 Michigan State University College of Education, 620 Farm Lane, Room 403, East Lansing,

MI 48824, USA

2 Hope College, Holland, MI, USA

3 Grand Valley State University, 401 W. Fulton, 476C DeVos, Grand Rapids, MI 49504, USA

123

J Educ Change (2016) 17:85–113

DOI 10.1007/s10833-015-9262-4

Introduction

A persistent issue in closing the achievement gap is improving the quality of

teaching and learning in urban schools. Many argue that improving urban schools

requires increasing the instructional capacity of teachers through job-embedded

professional development (PD), where teachers engage in collaborative, ongoing

dialogue around teaching and learning (Darling-Hammond et al. 2009; Heck and

Hallinger 2009; Horn and Little 2010). Such PD often relies on teachers assuming

formal roles as ‘‘teacher leaders’’ who guide this learning (Lieberman and Friedrich

2010; Yost et al. 2009). For this leadership to lead to improved instruction, however,

teacher leaders must skillfully engage in leadership practice that effectively changes

how their colleagues teach. Yet, the process by which teacher leaders create such

change is not clear in the extant literature. Thus, we conducted yearlong embedded

case studies of eleven urban teacher leaders working in teams to improve the

instruction of their colleagues by leading teacher learning around discussion-based

teaching—that is, by trying to help their colleagues better structure and lead

conversations among students. Integrating complex systems theory (Opfer and

Pedder 2011) with Kotter’s (1996) theory on leading organizational change, we

analyze how the embedded systems within which teacher leaders operate shape the

change actions they take and whether and how those actions change teaching

practice.

The challenge to improve urban schools

Movements to improve urban schools have been debated and mandated by policy

makers and business leaders for decades. Yet, as Payne (2008) asserts, ‘‘Most

discussion of educational policy and practice is dangerously disconnected from the

daily realities of urban schools’’ (p. 5). Such schools often lack resources such as

adequate funding, qualified teachers, and instructional leadership. Urban schools

also face high rates of student and teacher turnover, and students often come from

poverty-stricken homes. Payne argues that multiple social barriers (e.g., low

expectations, pessimistic views of new programs, and distrust between colleagues

and leaders) and micropolitical barriers (e.g., perceptions of favoritism and power

struggles) exist within urban schools that further hinder reform efforts. Although

schools may adopt the rhetoric of new programs readily, they often fail to

effectively meet the intent of such programs or adapt initiatives to their school

context. Similarly, popular reforms such as instructional coaching and decentral-

izing decision-making often fail because of power struggles between coaches and

school leaders and because teachers are often left out of decision-making processes.

Through all of these initiatives, Payne identifies teacher resistance as a central

problem in improving urban schools. Thus, it seems pertinent to consider the

teacher’s role in creating authentic change.

86 J Educ Change (2016) 17:85–113

123

Teacher leadership

Teacher leadership, in which teachers themselves generate and facilitate change, is

rooted in the teacher professionalism movement that began in the early 1980s and

continues today (Fairman and Mackenzie 2014; York-Barr and Duke 2004). Over

the past decade, the role of teacher leaders in school reform has become more

prominent in empirical research, and much of this research has posited that teacher

leaders are vital for successful school reform (Angelle and Schmid 2007; Crowther

et al. 2002; Frost et al. 2000; Katzenmeyer and Moller 2001; Murphy 2005; Valli

et al. 2006). The role of the teacher leader—what it is and how it is defined—is

varied, however, depending on the school context and the research. Yet, most

scholars agree that teacher leadership occurs within and outside classrooms to

influence school-wide instructional practice (Beachum and Dentith 2004; Katzen-

meyer and Moller 2001). Beyond role-specific duties or titles (such as department

chair or grade-level leader), teacher leadership rests with the agency of the teacher

to work with the principal, to build community, to support teachers, and to

determine, implement, or make manifest a school-wide vision for instructional

practice (Cranston 2000; Margolis and Huggins 2012; York-Barr and Duke 2004).

In reviewing the literature, York-Barr and Duke (2004) concluded that the

success of teacher leadership depends on interrelated, foundational conditions in

three areas: (a) school culture, (b) relationships, and (c) school structures. First,

researchers have argued that, for schools to exhibit positive change through teacher

leadership, they must have cultures that foster communication, collaboration, and

learning (Little 2006; Wood 2007). The principal must be open to and supportive of

teacher leaders, understand the teacher leaders’ work, and ensure they have a

prominent and visible role in developing the mission and values of the school

(Drago-Severson 2007; Little 2006; Mangin 2007; Wood 2007). Moreover, the

principal, teacher leader, and school faculty should work together to identify and

consistently uphold professional norms for collective learning and improved student

achievement and instruction. Secondly, teacher leaders need to build professional

and respectful relationships with colleagues through ongoing communication and

feedback that showcase their trustworthiness and instructional expertise. York-Barr

and Duke (2004) found that effective teacher leaders are generally seen as role

models, are respected by colleagues, and have leadership capacities. Teacher leaders

and principals also need to build positive relationships with one another, as

principals play a central role not only in developing teachers’ leadership skills, but

also in setting expectations and creating pathways for teacher leaders to succeed

(Mangin 2007). Finally, specific school structures that promote and support

effective teacher leadership include time for collaboration, shared leadership, and

embedded professional development (Drago-Severson 2007; Kardos et al. 2001;

Lampert et al. 2011; Little 2006; Paine et al. 2003). Although such structures

contrast with traditional hierarchical school structures and teacher isolation, which

are inherent in many schools (York-Barr and Duke 2004), when teachers have time

to discuss and plan instruction, analyze student work, and learn from others’

expertise, they can improve instruction and student learning (Chenoweth 2009;

J Educ Change (2016) 17:85–113 87

123

Drago-Severson 2007; Kardos et al. 2001; Little 2006). Shared leadership between

school leaders and faculty, such that faculty have a voice in decision-making

processes, also supports teacher leadership (Drago-Severson 2007), as does PD that

provides teachers with individualized learning opportunities connected to their

everyday instructional practice (Borko et al. 2008; Drago-Severson 2007; Little

2006). With the work of teacher leaders embedded in such PD, they can support

their colleagues as they promote valuable, engaging teacher and student learning

(York-Barr and Duke 2004).

The process of teacher leadership

Despite the breadth of research on the foundational conditions for teacher

leadership, this body of work does not present a complete picture of how teacher

leadership can and does improve instruction. That is, even when these conditions are

met, teaching and learning do not necessarily improve. There is a little-understood

teacher leadership process by which teachers take actions that lead to change in

their organizations. York-Barr and Duke (2004) identify three broad means of

influence by which effective teacher leaders can shape the work of individuals,

groups, and organizations. Those means of influence are broadly conceived and

include maintaining a focus on teaching and learning, establishing trusting and

constructive relationships, and interacting through formal and informal points of

influence. York-Barr and Duke identify the ultimate outcomes of such influence as

improved instructional practices and student learning. Yet, they do not articulate the

specific actions and tactics teacher leaders can take as they engage in those

relationships and interactions that would effectively change, rather than merely

influence, the instruction of other teachers. In distinguishing between these two

outcomes, we conceptualize influence as indirectly altering another’s practice by

informing their thinking in ways that shape what they do, whereas change is

intentionally propelling others to do some specific thing in a specific way that differs

from current practice.

In expanding on York-Barr and Duke’s work, Fairman and Mackenzie (2012) use

interviews with forty formal and informal teacher leaders to describe nine activities

in which teachers can influence instructional change, such as through collaborating

with peers or contributing to school improvement efforts. They position these

teacher leadership activities on a continuum from classroom-based to school-based.

More recently, Fairman and Mackenzie (2014) describe specific strategies these

same teacher leaders use to influence colleagues, such as by creating collegial

climates or building trusting relationships, and they provide examples of ways

teacher leaders have enacted these strategies. Through examining teacher leaders’

self-reports, Fairman and Mackenzie contribute to our understanding of the teacher

leadership process by delving more deeply into the actions individual teacher

leaders take. However, their findings rely on reflections from teacher leaders and

inferences about cause-and-effect relationships that may overstate the impact of

teacher leadership on ultimate outcomes. As with much of the research, Fairman and

Mackenzie’s conclusions rest on two assumptions: (a) that teacher leaders have a

88 J Educ Change (2016) 17:85–113

123

means to influence their colleagues’ work, and (b) that teacher leaders engage in

actions that lead their colleagues to change their practice. In the present study, we

examine these assumptions by analyzing videos of teacher leaders attempting to

produce change in their colleagues’ teaching and by using interviews and other

triangulated data to contextualize these change efforts in the embedded systems

within which teacher leaders function.

Theoretical framework

We draw on Kotter’s (1996) eight steps for leading organizational change as a

framework for classifying the tactics teacher leaders use when attempting to change

the practice of their colleagues. The change process, according to Kotter (1996),

begins with a sense that the status quo is not working. Outside factors, which in

schools might be low test scores, may serve as the impetus, but real change occurs

only when an internal sense of urgency motivates individuals to change what they

do (Step 1). For this to happen, individuals with power (e.g., administrators, formal

teacher leaders, influential teachers) take up the mantle of change and form a

guiding coalition (Step 2). This coalition leads initial change efforts by clearly

articulating the problem, developing a vision for the change process, and defining

feasible and focused strategies for enacting that vision (Step 3). The challenge for

the coalition is to ensure that individuals at all levels of the organization understand

and ‘buy in’ to the vision. In schools, coalitions might accomplish this by

championing a new instructional practice, trying it out themselves, and making it

central to their work with teachers. Their work also involves communicating the

vision in various modes and forms (e.g., faculty meetings, hallway/lunchroom

conversations, testimonials, etc.) and delivering a consistent message in ways that

appeal to the hearts and minds of teachers (Step 4). The goal is to embolden teachers

to try new ideas, convincing them to make the necessary sacrifices involved in

changing their instructional practices. As part of this work, the coalition provides

supports, such as resources (time, funds, and materials) and training, to empower

broad-based action toward the vision (Step 5). They make way for this action by

removing obstacles to the vision and confronting people who undermine change

efforts (intentionally or not). As the changes begin to take hold, the guiding

coalition focuses on creating and highlighting short-term wins that propel further

action (Step 6), and they turn their attention to producing more change by

acculturating new members, constantly revisiting the vision, and ensuring that all

decision-making relates directly to the change goals (Step 7). Throughout the

process, the guiding coalition operates with the full understanding that they will, at

some point, relinquish power to others as change begins to spread and new practices

become anchored in the culture of the school (Step 8).

To contextualize teacher leaders’ efforts to create change, we also utilize

complex systems theory (Opfer and Pedder 2011), which recognizes that teacher

learning is nested within complex systems that have varying levels of overlap and

influence. Lasting change can only take hold when extending beyond a guiding

coalition and becoming prevalent among many individuals in a school, each of

J Educ Change (2016) 17:85–113 89

123

whom possesses their own personal system of orientations toward a given reform—

such as their beliefs, understandings, and experiences with the reform. Achieving

change that permeates many systems within the school context is perhaps the most

complicated step in achieving true transformation. Opfer and Pedder (2011) assert,

‘‘Teacher learning tends to be constituted simultaneously in the activity of

autonomous entities (teachers), collectives (such as grade level and subject groups),

and subsystems within grander unities (schools within school systems within

sociopolitical educational contexts)’’ (p. 379). They frame these systems and

subsystems as ‘‘interdependent and reciprocally influential’’ (p. 379). They argue

that examining the nested systems in which teacher learning occurs sheds light on

‘‘the complex relationships between systems that promote and impede teacher

learning and instructional change’’ (Opfer and Pedder 2011, p. 379). As teachers are

asked to assume a leadership role, formal teacher leaders coexist within both the

leadership team and the teaching staff. In this unique boundary-crossing position,

teacher leaders may have a voice in decision-making and goal-setting, yet can

maintain their access and credibility with teachers, all of which may allow them to

play an important role in conveying the necessary sense of urgency to initiate and

propel change. From a complex systems perspective, we posit that teacher leaders

can link the visioning process and the implementation of new teaching practice

while also shaping the school’s operative culture across multiple systems (Opfer and

Pedder 2011). Once the change process begins, teacher leaders could contribute

further by connecting systems as they communicate a consistent vision and engage

in the learning process with their colleagues. As we have hinted at here, the present

study seeks to situate Kotter’s (1996) eight steps for leading change within the

complex systems that frame teacher leaders’ work as a way to understand the

process of teacher leadership.

Professional development for teacher leaders

This study accompanied our work as university-based PD providers in a 4-year

program with 28 high-poverty urban charter schools in a large Midwestern city.

Two or more teachers at each school were placed in formal teacher leadership

roles—as professional learning community leaders (PLC leaders) or mentors for

novice teachers—to build a school-wide culture of professional inquiry around

discussion-based teaching. Schools were invited to participate in this PD as part of

their required work in a large federally funded grant initiative led by the state

charter association. Their participation in the grant provided multiple years of free

professional development, along with school resources and stipends for teacher

leaders. Our PD focused on developing the practice of teacher leaders who could

facilitate inquiry-oriented PLC meetings that enhanced the quantity and quality of

professional dialogue among teachers. At the same time, we prepared mentors to

work one-on-one with beginning teachers to further facilitate professional dialogue

and teacher learning. At the outset of the program, we introduced school principals

to a rubric (Stanulis et al. 2011) to help select their PLC leaders and mentors. Rubric

criteria included elements of Dewey’s (1933) characteristics of educative teachers,

90 J Educ Change (2016) 17:85–113

123

including being wholehearted (approaching teaching with joy and connecting

content and students in meaningful ways), trustworthy (opening practice to others

and valuing conversations with colleagues about teaching and learning), and

openminded (being open to learning and eager to try new ideas; being open to

reflection and analysis of one’s teaching). The rubric also integrated York-Barr and

Duke’s (2004) foundational conditions that teacher leaders be respected as teachers,

be learning oriented, and have leadership capacities. Depending on the size of their

school, principals selected one or two PLC leaders and one to four mentor teachers.

Because principals are foundational to school change (Grissom and Loeb 2011), we

included the principals in three PD sessions each year, building on the initial session

where we discussed selection of mentors and PLC leaders. The content of this PD

focused on ways to develop and support a school culture that supports teacher

learning, and we provided modeled examples of principles and practices of an

effective inquiry-focused PLC meeting. Principals were also updated on themes of

the mentor/PLC leader sessions and regularly met with their PLC leader and

mentors as a school team to plan next steps for enacting instructional change in their

school.

In the 2012–2013 school year, PLC meetings and mentoring both centered on

promoting an inquiry-based learning climate around discussion-based teaching.

Discussion-based teaching is an instructional strategy for improving student

learning, as it (a) involves students in meaning making while actively listening

and voicing ideas; (b) requires students to provide relevant evidence from materials

being discussed; (c) promotes student linking, in which students agree with,

disagree with, and/or extend their peers’ contributions; and (d) provides opportu-

nities for students to communicate orally and in writing to improve critical thinking

and comprehension (Jadallah et al. 2010; Matsumura et al. 2010; Stanulis et al.

2014). All teachers engage students in some talk, but few learn to lead the kinds of

classroom discussions that are rigorous and lead to critical thinking (Gambrell and

Almasi 1996). Learning to lead effective discussions involves knowledge of

individual students, textual material, and ways of reading and scaffolding

participation to facilitate these kinds of discussions (Stanulis and Brondyk 2013).

We chose to focus on discussion-based teaching in our PD with teacher leaders

because teachers benefit from focused attention on one area of instructional

improvement and because learning to lead classroom discussions is a core practice

linked to effective teaching (Ball et al. 2009; Grossman et al. 2009). We undertook

the present study with three school teams from our PD to understand how teacher

leaders engaged in these instructional improvement efforts at their schools.

Research questions

To understand the process by which teacher leaders in our PD program attempted to

change their colleagues’ instructional practice around discussion-based teaching, we

set out to answer two research questions:

J Educ Change (2016) 17:85–113 91

123

1. What change tactics do the leadership teams and individual teacher leaders use

when attempting to change the teaching practice of their colleagues, and how do

they use them?

2. How do the structural and cultural facets of the systems within which teacher

leaders are situated, including the leadership teams in which they are embedded,

promote and impede their efforts to create change?

Methods

We developed embedded case studies (Yin 2009) of leadership teams in three

schools participating in our PD. The overarching cases were the teams, each

consisting of a principal and 2–5 teacher leaders, and the eleven teacher leaders

across the three schools constituted embedded cases. This embedded design enabled

us to examine how the members of each team functioned collectively and separately

in their efforts to change the teaching practices of their colleagues and how various

embedded systems shaped teacher leadership in three different schools.

Site and participant selection

Following the first year of PD (2011–2012), we used purposeful theoretical

sampling (Patton 2002) to identify three schools in which we perceived a high

likelihood that teacher leadership was impacting school-wide instruction. Based on

our interactions with principals and teacher leaders at all 28 participating schools,

we targeted three sites in which the teacher leaders at that time appeared to have

strong leadership capacity and a learning orientation and where the principals

appeared to support teacher leaders by devoting ample time and resources to their

work. In these ways, these schools appeared to meet York-Barr and Duke’s (2004)

foundational conditions for effective teacher leadership and thus be suitable loca-

tions to study the process of teacher leadership in an urban setting.

Table 1 provides an overview of the three overarching school cases—Spruce,

Maple, and Dogwood Academies (pseudonyms)—and the eleven embedded teacher

leader cases (shown in bold). The number of teacher leaders at each school ranged

from two to five, and all participating leaders were female. Table 1 also provides

size, demographic, and academic performance data for the schools. All three were

charter schools in a high-poverty urban area in the Midwest and had student bodies

that were 94–99 % African American, ranging in size from 355 to 575 students.

Despite similar demographics and test scores, however, these schools ranked quite

differently in the 2011–2012 statewide ranking of schools, which is based on test

scores, growth over time, and within-school achievement gaps. As of 2012, Spruce

and Dogwood were both in the upper 40th percentile, while Maple was struggling

by comparison at only the 7th percentile.

We selected these schools based on our assessment of the principals, teacher

leaders, and school contexts before our study began. At Maple and Dogwood

Academies, there were high levels of consistency in school leaders, teacher leaders,

92 J Educ Change (2016) 17:85–113

123

teaching staffs, and student bodies from the year of selection to the year of study.

However, there were many interim changes at Spruce Academy. First, Spruce’s

K-12 campus divided into four separate schools—with kindergarten, elementary,

middle, and high schools. The principal, Sophia, became superintendent of all four

schools, and administrative interns were placed in the elementary school, the site for

our study. The student and teacher populations also tripled, bringing in many new

teachers who were recent college graduates. In addition, Spruce’s original PLC

leader moved to the high school, leaving the kindergarten and elementary teachers

to work with a new PLC leader, Stacy, who joined our study. As such, the Spruce

Academy we studied was contextually quite different from the one we recruited.

Although our intention was not to study a school undergoing such transition, our

interest in the impact of context on teacher leadership ended up making this an

informative site.

Data collection

We collected data during the second year of our PD (2012–2013). To inform our

first research question regarding change tactics, we collected seventeen self-

captured videos of the teacher leaders engaging in leadership work. PLC leaders

digitally video-recorded themselves facilitating or co-facilitating two whole-staff,

Table 1 Overview of the three overarching school cases (with embedded cases noted in bold)

Spruce

Academy

Maple

Academy

Dogwood

Academy

Leadership team

Principal Sophia Marie Donna

PLC leader(s) Stacy Michelle Daphne

Morgan

Mentor(s) Sarah Maggie Debbie

Melinda Dawn

Madeline Dianne

Size and demographics 2012

Grades served K-5 K-5 K-8

Number of teachers 34 30 16

Number of students 575 510 355

% African American 94 99 99

% Economically disadvantaged 84 83 70

Academic performance data 2012

% proficient/advanced in reading (5th

grade)

59 47 66

% proficient/advanced in math (5th grade) 15 5 15

State rankinga 74th percentile 7th percentile 63rd percentile

a State ranking is based on test scores, growth over time, and within-school achievement gaps

J Educ Change (2016) 17:85–113 93

123

hour-long PLC meetings—one in fall and one in winter or spring. Mentors recorded

themselves engaging in one-on-one mentoring conversations with beginning

teachers. Three mentors video-recorded one mentoring session, and four mentors

recorded two sessions—one early and one later in the year. Mentoring sessions

ranged from 8 to 30 min. All teacher leaders wrote reflections on their videos and

discussed the videos and reflections in a formal coaching conversation with a

member of our team. The videos, written reflections, and notes from the coaching

conversations all served as data on the change tactics teacher leaders used and how

they understood their efforts to lead instructional change.

To inform our second research question on the influence of embedded systems on

teacher leadership, we conducted thirty-four interviews with the PLC leaders,

mentors, and principals across the year. Interviews ranged from 30 to 60 min and

were recorded and transcribed. Nine teacher leaders were interviewed three times

each at the beginning, middle, and end of the year. The two PLC leaders at Maple

Academy were interviewed together each time since they shared the role. Two

mentors were only interviewed twice due to health issues. School principals were

interviewed once (1 principal) or twice (2 principals) as their schedules allowed. All

interviews followed a semi-structured protocol that sought to identify the

individual’s understanding of their and others’ leadership experiences, their

responsibilities and challenges as a leader, and how they saw the team, school,

and broader contexts as impacting their efforts to change instruction.

To triangulate our understanding of the leadership practice and contexts at these

three sites, we collected three additional forms of data. During three visits to each

school, we collected artifacts from leadership activities, including agendas from

PLC meetings, samples of teacher work products from PLC meetings (posters,

lessons plans, etc.), and copies of materials distributed during PLC meetings and

mentoring sessions. At the end of the year, we also surveyed the full teaching staff

at each school to assess teachers’ perceptions of discussion-based teaching in their

classrooms (the primary focus of the teacher leaders’ work) and the quantity and

quality of professional dialogue among teachers in the school. Surveys were

administered during staff meetings and were completed by at least 75 % of the

faculty at each school. Finally, we used school and state websites to collect data on

student achievement and staff attrition at each school.

Data analysis

Data analysis occurred in three phases. In Phase I, as data were collected, we used

our theoretical frameworks to generate descriptive codes and coded the thirty-four

interviews (Miles and Huberman 1994). In Phase II, following data collection, we

met as a six-person research team over 2 months to generate individual case reports

(Yin 2009) for each teacher leader by collectively viewing the seventeen video-

recordings, examining supplemental data, and discussing the practice of each

teacher leader to identify where and how she used Kotter’s (1996) change tactics.

The member of our team who coached each teacher leader during the PD added

information from their notes and personal interactions. We also reviewed interview

excerpts from the teacher leaders and principals, discussed our interpretations, and

94 J Educ Change (2016) 17:85–113

123

added additional notes to the case reports. We then summarized our impressions of

all eleven teacher leaders in an analytic matrix (Miles and Huberman 1994)

organized by Kotter’s eight change tactics.

At this point, we moved into Phase III and considered the three leadership teams

as cohesive units of analysis. Here we integrated Opfer and Pedder’s (2011) theory

with Kotter’s (1996) to consider aspects of the change process. Questions that

propelled our analysis include: Who drives change? How do various members of the

team shape the change process? Is there a sense of urgency that change must

occur? Does the leadership team have a strong vision, and is it clearly and

regularly communicated to others? How do school leaders support or impede the

process? How does the school context impact this work? We drafted one-page

interpretive statements to describe how each team attempted to change instruction,

and we created concept maps to capture the embedded systems and team dynamics

at each school (Maxwell 2005). Figure 1 shows a generic concept map to illustrate

our graphic conceptualization. The left side of the model denotes how complex

systems interact in a process that works toward a particular product on the right side

of the model, such as sustainable school-wide change in instruction. The gray circles

at the far left represent two systems that were constant for our three cases, as all

teacher leaders received the same PD and worked in similar contexts. The center of

the graphic shows the various players who influenced change. We conceptualized

that teacher learning takes place within the arc of the model and is influenced by the

school leader, who has the ability to enable or constrain teacher learning by

influencing the school context for teacher learning (such as by scheduling time and

allocating resources). Teacher leaders also influence teacher learning within the arc

as they interact with one another and their colleagues. In our analyses, we modified

this model for each school to depict how these complex systems intersected in ways

Sustainable School-wide

Change

PD

Colleagues Teacher Leaders

CONTEXT

Process Product

Fig. 1 Generic concept map denoting how embedded systems can shape the teacher leadership change process in a school

J Educ Change (2016) 17:85–113 95

123

that promoted or constrained instructional change at each site. As a final step, we

looked across our analytic tools to identify our overall findings regarding the

process of teacher leadership and how the school contexts and leadership teams

shaped those processes.

Findings

Given Opfer and Pedder’s (2011) complex systems theory, we theorized that teacher

leaders’ efforts to create change would be influenced by the embedded systems

surrounding their work. This could not have been more true. We found that change

efforts among these eleven teacher leaders were overwhelmingly shaped by a

multitude of systems, including the contexts of their leadership teams and schools,

their individual PD experiences, their personal orientations toward teacher

leadership and those of their colleagues, and external factors such as being in a

large city and being charter schools. Using the lens of Kotter’s (1996) eight steps for

leading change, we also inferred that some teacher leaders made key missteps early

in the change process—particularly in the initial four phases of establishing

urgency, creating a guiding coalition, developing a vision, and communicating that

vision—that limited their ability to change their colleagues’ practice. Examining

change tactics and embedded systems simultaneously, we identified key ways in

which complexities within the interlocking systems—particularly the leadership

teams and school contexts—greatly influenced whether, and if so how, teacher

leaders individually and collectively enacted Kotter’s first four steps toward change.

In the end-of-year survey of the full teaching staff, teachers at all three schools

reported attending regular PLC meetings with genuine participation among their

colleagues, and most teachers at all three sites agreed that mentors helped new

teachers think about student learning. We also found, however, that teachers in the

three schools reported using discussion-based teaching strategies with different

frequencies at the end of the year. At Spruce and Maple Academies, only 22–25 %

of teachers reported having students actively talk and participate in class more than

three-quarters of the time, and only 14–18 % of teachers reported having students

talk with a peer or in groups for more than 60 min a week. By contrast, these

percentages at Dogwood were 66 % for students actively talking and participating

more than three-quarters of the time, and 58 and 42 % for students spending more

than 60 min a week talking to a peer or with a group, respectively. Although we do

not have baseline data to assess whether the frequency of discussion-based teaching

actually changed over the year, our qualitative data strongly support the conclusion

that the high volume of discussion-based teaching reported at Dogwood was due in

large part to the teacher leaders. In particular, our analyses reveal that the nature of

teacher leadership at Dogwood was much more systematically focused on the goal

of increasing discussion-based teaching than that at either Spruce or Maple, and the

embedded systems at Dogwood greatly supported this teacher leadership work in

ways we did not find at Spruce and Maple.

To illustrate how embedded systems shaped teacher leaders’ efforts to create

change, we present our findings on each of the three cases below. We begin with

96 J Educ Change (2016) 17:85–113

123

Dogwood Academy, the one school in which teacher leaders successfully enacted

Kotter’s (1996) first four steps for leading change and ultimately appeared to create

instructional change among their colleagues. We start here to lay the foundation for

how embedded systems can support the teacher leadership process. We then present

the Maple and Spruce cases—in which Kotter’s framework suggests that teacher

leaders took missteps early in the change process and thus ultimately had less

impact on the implementation of discussion-based teaching—to illustrate two ways

in which embedded systems can hinder or undermine the teacher leadership process.

Dogwood Academy: Embedded systems supporting the teacher leadership process

As a small charter school on a busy city street, Dogwood Academy was comprised

of a tight-knit, sixteen-person teaching staff of primarily African American women.

Figure 2 shows our conceptualization of how embedded systems supported the

change process at Dogwood. Central to instructional change was the strong and

focused leadership of the PLC leader, Daphne (shown by the gray circle at the base

of the teacher learning arc), who had been teaching at Dogwood for 6 years. Her

principal Donna remarked of her, ‘‘She’s personable and humble so that you know

ego doesn’t get in the way…. She is very inclusive of the staff. She values their

ideas. She wants them to contribute… She is the biggest cheerleader in the group.’’

Although our PD did not explicitly address Kotter’s (1996) eight steps for leading

change, Daphne intuitively enacted the first five steps and was well positioned to

move into the later stages of the model.

As the foundational step for change, Daphne did what PLC leaders at the other

schools did not: She created a sense of urgency around the need for more

discussion-based teaching. She did this by linking discussion-based teaching to an

overall vision for change that she labeled ‘‘Bridging the System,’’ and which she

linked to the particular needs of Dogwood’s urban students whom she argued

needed consistency. Daphne described this vision in an interview:

Discussion-based Teaching

Mentor

Mentor

Mentor

PD

Context

PLC Leader Colleagues

(16) PLC

Leader

Fig. 2 Change process at Dogwood Academy

J Educ Change (2016) 17:85–113 97

123

What is resonating throughout our building right now is just bridging our

system and everybody is talking about it…. It is our system, it’s not, you

know, 4th grade scores…. Our students should have a lot more consistency,

and we’re doing them a disservice [when we’re inconsistent]. So our goal is to

figure out where are we losing, why are we losing, and how can we stop losing

it…. Everybody understands that it is the system and we have to work on our

system in order for our students to succeed.

As part of trying to increase consistency throughout the school, Daphne’s whole-

school PLC meetings heavily emphasized uniform approaches to discussion across

the grade levels. In a spring PLC meeting, Daphne asserted to her colleagues, ‘‘If

something works, why shouldn’t it be modified for all grades?’’ During that

meeting, Daphne had two teachers share best practices in discussion-based teaching,

and teachers met in grade levels to modify those practices for each grade. In creating

urgency around a vision of ‘‘bridging the system,’’ Daphne drew on the needs of

Dogwood’s population of urban students, thereby integrating the local context, and

urged her colleagues to consider their responsibility to scaffold student learning

over multiple years.

This compelling, urgent focus on discussion-based teaching also appeared in the

practice of two mentors, Debbie and Dianne. In a representative exchange, Debbie

and a mentee debriefed a lesson. The mentee reflected, ‘‘I suppose it’s my job to

make sure [the students] know prior to playing the game that the most important

thing is student discussion. And I might not have set that up quite as well as I could

have.’’ Debbie concurred, and they considered strategies for more successfully

structuring group conversations. Despite the consistent focus among Daphne,

Debbie, and Dianne (as shown by overlapping circles in the model), the third

mentor, Dawn, did not contribute to increasing discussion-based teaching. As a

counselor, rather than a teacher, Dawn’s mentoring interactions tended to center on

students’ psychological needs and rarely addressed instructional strategies. The

other three teacher leaders all reinforced one another’s messages that discussion was

central to the vision of effective teaching at Dogwood Academy.

School leader Donna reinforced the vision and functioned as a co-visionary with

Daphne. She also supported the change process by creating structures to support

teacher learning (represented on the model by the arc with arrows), and she

regularly attended our PD sessions for school leaders, actively participating and

sharing her inclusive vision of teacher leadership with other principals. Daphne

noted of Donna, ‘‘She models what she hears from the large PLCs…. That’s the way

she carries her meetings…. We all have input; she listens to us. So I know that she

feels that student talk is important, I know it.’’ Despite modeling the vision herself,

Donna enabled Daphne to guide PLC meetings. Donna explained,

I don’t always stay for the entire [meeting] as long as I have had a

conversation with Daphne beforehand and we know what the agenda is. She’s

very good about sharing what they’re going to do and saying, ‘Is this in line

with what you would like to see?’ And a lot of times I say, ‘Well we’re

working this together. It’s not so much what I want to see you do as long as

we’re all on the same page.’

98 J Educ Change (2016) 17:85–113

123

Enacting a unified and clearly communicated vision for instructional change, the

guiding coalition of three teacher leaders and the principal were able to move into

Kotter’s (1996) fifth step for leading change by empowering broad-based action.

Dogwood’s collegial culture of safety and shared expertise supported frequent

conversations about teaching practice that enabled broad-based action to take hold.

Indeed, everyone we interviewed at Dogwood described a highly collegial

professional context. Mentor Debbie described, ‘‘This is like a family, and we

have one another’s back…. This is a great, great environment.’’ In one video,

Debbie’s mentee remarked, ‘‘You know, Debbie, I appreciate you and some of the

other middle school teachers…. I appreciate the help you’ve given me and just the

acceptance [and] the professional relationship that we’ve had over the last eight,

nine months.’’ This highly collegial tone permeated the interactions we witnessed in

the videos, our own interactions with this leadership team, and our campus visits to

Dogwood. In this context, the teacher leaders were able to position themselves as

co-learners with colleagues and influence broad implementation of discussion-based

teaching.

As depicted in Fig. 2, the evidence illustrates that the structure and culture of the

leadership team and school context at Dogwood created the space and conditions to

support teacher-leader-driven change. Discussion-based teaching was clearly a

common focus of teacher learning at Dogwood at all levels throughout the year, and

conversations in the videos along with the end-of-year survey results suggest that

discussion-based teaching encompassed the practice of many of the sixteen

colleagues in the school by the end of the year.

Maple Academy: Embedded systems pulling in many directions

At the outset, Maple Academy appeared to have many key elements that we

envisioned would support instructional change through teacher leadership. The

school had a strong principal, Marie, who was popular with her staff and highly

invested in the instructional quality of her school. Marie always attended our PD,

volunteering for her team to share ideas and smiling proudly as staff members

shared what was happening in their school. The staff of primarily young white

women also had very low turnover, with all thirty of the 2012–2013 teachers having

taught at Maple the year prior. Two experienced Maple teachers, Michelle and

Morgan, shared the PLC leader role, and worked as part-time administrators in tight

collaboration with Marie. The three mentors were experienced teachers, whom

Marie described as ‘‘my strong classroom leaders, grade level team leaders.’’

However, we ultimately found that the embedded systems at Maple were relatively

rigid and disconnected, pulling the leadership team in many directions and

seemingly undermining the teacher leaders’ ability to create change. In Fig. 3, we

illustrate our conceptualization of the change process at Maple. The most noticeable

feature of this graphic is the number of arrows pointing to various initiatives. Video

analyses of PLC meetings and mentoring sessions at Maple revealed a fairly diffuse,

disconnected vision for instructional change due to the large number of initiatives

and emphases during the 2012–2013 school year. These included implementation of

Common Core State Standards, preparing for the new Smarter Balance assessment,

J Educ Change (2016) 17:85–113 99

123

monthly trainings on Classroom Instruction that Works, Daily Five literacy centers,

posting objectives on the board, cooperative learning, small-group Data Analysis

Teams, meeting the goals outlined in the School Improvement Plan and Maple’s

charter, and discussion-based teaching. With so many simultaneous initiatives each

being treated as a separate endeavor, discussion-based teaching was only one

strategy among many (hence the small PD overlap in Fig. 3). Not only could we not

locate discussion-based teaching within the school’s vision, we could not identify a

singular clear, unified vision for effective teaching at Maple. We further noted that

the list of initiatives underway at Maple were context independent; they were not

framed as being responsive to the particular learning needs of Maple students. This

stood in stark contrast to Daphne’s assertion that Dogwood students needed

consistency across grade levels.

Also prominent in Fig. 3 is the thick, all-encompassing arc representing principal

Marie’s top-down managerial style. In an interview, Marie listed ‘‘non-negotiables’’

for teachers, including posting instructional objectives and using cooperative groups

and ‘‘Daily Five’’ literacy centers. In many ways, the teacher leaders were Marie’s

enforcers. Mentor Madeline relayed,

Maggie had an issue where Daily Five was really not even being done at all

[by a mentee]. It was called Daily Five, but it wasn’t [really Daily Five], and

so she went to the principal and said, ‘How important is it that this teacher is

doing this?’ And the principal said, ‘Well, it’s very important.’ And she said,

‘Well, it’s a huge problem because it’s not happening at all.’

Through such anecdotes, it became clear that urgency at Maple was centered on

using the teaching strategies Marie advocated and that Marie was the driving force

for teacher learning. As a group, the principal and teacher leaders at Maple had a

relatively large team of six to serve as a guiding coalition. Yet, because the

instructional agenda came from the principal and was not owned nor shaped by the

teacher leaders, they had little autonomy and few opportunities for leading change.

PLC Leader

PLC Leader

Mentor

Mentor

Mentor

Posting instructional objectives

Cooperative learning

Daily Five

Common Core State Standards

Preparing for Smarter Balance

Instruction that Works

Teacher data analysis teams

Discussion-based Teaching

PD

Context

Colleagues (34)

Fig. 3 Change process at Maple Academy

100 J Educ Change (2016) 17:85–113

123

Thus, in Fig. 3, none of the teacher leaders are shaded because none took the lead in

creating a vision for or movement toward discussion-based teaching. Although a

few mentors mentioned discussion-based teaching when talking with mentees, those

practices got lost in the myriad of initiatives and, therefore, never took hold as a

means to school-wide change.

One area where teacher leaders at Maple did have an opportunity to lead change

was in communicating a vision, Kotter’s (1996) fourth step. Here, the PLC leaders

in particular possessed leadership opportunities as they ran PLC meetings and

attempted to guide their colleagues’ understanding of discussion-based teaching and

other initiatives. Through this communication, they had the potential to identify

common threads across various initiatives and weave them together in a way that

could solidify a change vision. However, Michelle and Morgan tended to preserve

the disconnected nature of the initiatives by talking about them separately rather

than synthesizing them into a cohesive vision. Michelle acknowledged, ‘‘I feel like

it’s hard to get in everything especially, you know, with all the other things that

we’re trying to do here. So, we’re trying to get as much as possible in without being

overwhelming.’’ Mentors reported using a written list of expectations. Madeline

described, ‘‘We have a checklist of objectives we’re supposed to discuss each

month, so we kind of go through those things. And then otherwise we talk about the

culture of talk and partner work and different ways to incorporate talk.’’ In these

ways, even though the teacher leaders had an opportunity to lead through

communication, they did not experience the autonomy to engage in such leadership.

Spruce Academy: Embedded systems lost in transition

We found that the school context at Spruce Academy was not well suited for

instructional change during the 2012–2013 school year due to the transition to

separate campuses, the tripling of the staff, the hiring of seventeen new teachers

(primarily young white women and men fresh out of college), and the promotion of

the principal to superintendent. In the midst of all this change, the embedded

systems surrounding teacher leadership were compromised and struggling to

become re-systematized. In large part due to this shifting context, the change efforts

of the two teacher leaders at Spruce were relatively disjointed and ineffective for

fostering wide implementation of discussion-based teaching across classrooms.

Prior to the transition, principal Sophia, mentor Sarah, and the original PLC leader

attended our PD sessions together and planned as a team. Sophia attended our

sessions for school leaders, facilitated leadership team conversations, and provided

regular release time for mentoring and two early-release days for students each

month so that teachers could work in PLCs. However, in this second year, with an

expanded role, Sophia was rarely present at the PD, left two administrative interns

to oversee the elementary building, and reduced release time for mentors and PLCs.

This new administrative ‘‘hands off’’ approach is represented in Fig. 4 by the wide

arc, denoting that the school leader did very little to create a context for teacher

learning. In her second year as a mentor, Sarah mentored eight beginning teachers,

far more than we recommended, in addition to teaching her own class. Stacy took

J Educ Change (2016) 17:85–113 101

123

over early in the 2012–2013 year as the new PLC leader at the elementary campus,

which meant she missed more than a full year of our PD.

It is within this changing, taxed context that we asked Sarah and Stacy to promote

discussion-based teaching as an instructional change agenda for their fellow teachers.

In examining Stacy’s efforts as the new PLC leader, we found that she did not

employ any of Kotter’s (1996) strategies for creating change. In regards to urgency,

which Kotter positions as the first step toward change, we found that Stacy’s sense of

urgency was centered on acclimating the large number of new teachers. As such,

Stacy prioritized building a professional community with high morale over

promoting any specific teaching practice, and she emphasized what she saw as the

needs of new teachers, including ‘‘things that… either they’re already implementing

in their classroom or they can pretty quickly implement.’’ At the end of the year,

Stacy reflected, ‘‘I think any changes have been better morale among the teachers.’’

Although discussion-based teaching was included in some PLC meetings, Stacy

presented the strategies in the form of ‘‘tips’’ for teachers, rather than research-based

practices for enhancing student learning. In one PLC meeting, Stacy explained,

Something [the PD team] mentioned the other day, and I want to do this

myself, is set your phone out during your lesson and put it aside and record

your voice…. How much are you talking verses your students talking? … I

think that was a really great idea and I want to do that myself. Just press

‘record’ so I can hear my voice and analyze and look back and reflect on

myself. How much talking am I doing as opposed to my kids? So that’s a

really good idea.

Like much of what occurred in Stacy’s PLC meetings, this excerpt illustrates a

casual yet upbeat attitude toward the ideas from our PD and a sense that her role as

PLC leader was to collect ‘‘tips’’ and pass those on to her fellow teachers. She did

Acclimating New Teachers

Discussion-based Teaching

BTs (8)

Context

Mentor

PD

PLC Leader

Colleagues (26)

Fig. 4 Change process at Spruce Academy

102 J Educ Change (2016) 17:85–113

123

not create a sense of urgency for teachers to develop expertise in discussion-based

teaching as a way to increase student learning. We also found some evidence that

Stacy was still in the process of developing her own understanding of discussion-

based teaching, as she sometimes conflated different practices in her PLC meetings.

For example, she regularly interchanged the terms ‘‘whole class discussion’’ and

‘‘whole class instruction.’’

Amidst Stacy’s focus on acclimation and her still-developing understanding of

discussion-based teaching, we found little evidence of a school-wide vision for

instruction at Spruce. In part, Stacy’s hesitancy to formulate a vision appeared to be

rooted in her personal orientation toward leadership and her interest in remaining

just one of the staff. She commented, ‘‘One of the things I still struggle with, from

time to time, is maintaining a level of leadership without making others feel I ‘know

all.’… I try to be as low-key and relatable as possible.’’ Reflecting this orientation,

Stacy closed a PLC sharing session among teachers by saying, ‘‘These are some

really good ideas and what I’ll do is copy all of these down and email them to

everyone, so if there’s one that you want to try in your classroom, you can do that.’’

It was clear from Stacy’s videos and interviews that her PLC meetings primarily

consisted of teachers sharing ideas without any criteria for those ideas or any central

thread connecting different practices around a particular vision of teaching. In this

way, Stacy did not appear to possess a vision for change and seemed to resist the

idea that she could be a visionary for Spruce.

In contrast to Stacy, we found that Sarah emphasized discussion-based teaching

more regularly and revealed stronger understanding of the student learning benefits.

Figure 4 includes divergent arrows to indicate that the two teacher leaders were

working toward different goals and so not aligned in pursuit of a common vision of

good teaching. Sarah as the mentor (shaded gray) was the driver toward discussion-

based teaching with her small group of beginning teachers (denoted by the BT circle

as a subset of the overall colleagues). Stacy, as the PLC leader, was more concerned

with acclimating new teachers as she engaged with the staff at large. Although

Sarah also had this goal, she embraced discussion-based teaching as a means for

acclimation. Just the same, we did not sense great urgency in Sarah’s mentoring, as

her tone was more soft and suggestive. In interviews, however, Sarah reported

observing and celebrating her mentees’ use of discussion-based teaching, suggesting

a more intentional focus than her mentoring tone revealed. She noted of one mentee,

‘‘The Spanish teacher, I love him to pieces because he actually is using ‘turn and

talk’ that I do with my kids…. They’re now sitting with each other and [having]

conversations with each other in Spanish and practicing polite mannerisms.’’

One clear issue impeding teacher leadership for change at Spruce was the small

number of teacher leaders. Kotter (1996) argues that change requires a guiding

coalition of powerful leaders who will collaborate on steering the change. Unlike at

Dogwood and Maple, where there were four or five teacher leaders and an on-site

principal, Spruce had only two teacher leaders who were functioning with very little

administrative oversight. In this way, Spruce’s guiding coalition appeared to be too

small, particularly compared to their staff of 34 teachers. In some ways, Sarah’s

cadre of eight mentees became something of a small coalition, meeting occasionally

to review discussion-based teaching. However, this coalition was not particularly

J Educ Change (2016) 17:85–113 103

123

powerful given their status as primarily new teachers, and they were not a force for

change throughout the school. Without a real champion for a school-wide vision

centered on discussion-based teaching and without a sense of urgency around the

need for such a vision, instruction at Spruce Academy did not broadly appear to

integrate discussion-based teaching during the 2012–2013 school year beyond the

small cohort of Sarah’s mentees. In this case, the transitional context of the school

clearly played a role in the minimal impact of teacher leadership.

Discussion

Looking across these cases, we gain insight into the process of teacher leadership as

it occurs within embedded systems and how those systems support, direct, or

impede instructional change. This study is not meant to be an endorsement or

critique of any particular school or school leader, as these school settings are much

more complex than what we can account for through only two theoretical lenses.

Rather, we provide comparisons across the schools in an effort to provide insight

into how schools that would like to affect change through teacher leadership might

learn from examples of schools and teacher leaders in similar contexts. The findings

from this study suggest that when teacher leaders work within networks of

supportive embedded systems, they can develop and drive change towards an

instructional vision that is clear and reinforced. However, when teacher leaders

work in environments that are disconnected or compromised, their ability to

influence or change their peers’ instruction is highly limited. Below, we discuss five

embedded systems that we found to impact the teacher leadership change process:

the teacher leader’s personal orientations toward leadership, the school principal’s

orientations toward leadership, the leadership team, the school context, and the local

context outside of the school. Throughout this discussion, we consider Kotter’s

(1996) first four steps for leading change—establishing urgency, creating a guiding

coalition, and developing and communicating a vision. Although these four steps

might oversimplify the complexity of leading change within dynamic organizations,

they help us begin to illuminate how various systems can impact whether and how

teacher leaders engage in initial steps toward creating change.

The teacher leader’s orientations toward leadership as a system

Critical components of teacher leadership included the teacher leader’s beliefs,

language, prior experiences, and knowledge base, which collectively constituted

their orientations toward leading and served as individual-level subsystems within

schools and leadership teams (Opfer and Pedder 2011). Certainly, not all of these

teacher leaders believed that they were, or could be, change agents, and such self-

perceptions impacted the level of boldness in their language and assertions. This

was most profoundly evident in the contrast between PLC leaders Daphne and

Stacy. Whereas Daphne tended to make strong assertions (e.g., ‘‘Our students

should have a lot more consistency.’’ ‘‘We have to work on our system.’’), Stacy’s

language was much more tempered (e.g., ‘‘If there’s one you want to try… you can

104 J Educ Change (2016) 17:85–113

123

do that.’’ ‘‘So, that’s a really good idea.’’). Given that Kotter (1996) positions

creating a sense of urgency as the first step in leading change, we surmise that the

differences in the linguistic tones of these two teacher leaders had differing impacts

on how compelled their colleagues felt to try new practices in their classrooms.

As prior research has noted (Stein and Nelson 2003; Timperley 2005), we also

found that teacher leaders’ depth of knowledge of the teaching practices they were

promoting influenced their change efforts. In delivering our PD, we felt we

presented our participants with adequate information on the rationale for and most

effective implementation of discussion-based teaching. However, in watching

eleven teacher leaders talk about these practices in the videos of PLC meetings and

mentoring sessions, we questioned the level of understanding among some of our

participants and reconsidered our own assumptions about the background knowl-

edge of principal-identified teacher leaders. Similarly, we identified a critical need

for teacher leaders to possess a substantial knowledge base about instructional

leadership and strategies for leading change. In training teacher leaders, we focused

on developing understanding of professional learning communities, mentoring, and

the types of collaborative practices that support teacher learning. What we failed to

consider was the need to also prepare teacher leaders to understand organizational

change and strategies for driving change among their peers—a process that turned

out to require much more assertive leadership and purposeful visioning than we

anticipated. As a result, teacher leaders at Maple and Spruce did not take action

toward leading change around discussion-based teaching that was as purposeful as

that taken at Dogwood.

The school principal’s orientations toward leadership as a system

Kotter (1996) argues that organizational change requires change agents to create a

guiding coalition of powerful leaders who will collaborate to take action on steering

the change. It is within this guiding coalition that teacher leaders can become

instructional change agents if principals provide them with appropriate amounts of

autonomy and support (Stein and Nelson 2003). As prior research has shown, the

extent to which teacher leaders have autonomy to actually engage in leadership lies

in great part with the principal (Mangin 2007; York-Barr and Duke 2004). By

examining how principals’ orientations toward leadership interact with other

systems to shape the teacher leadership process, we found that when the principal

made room for the voices of teacher leaders and was a member—as opposed to the

leader—of the guiding coalition, those teacher leaders effectively drove school-

wide change. Yet, when the principal did not relinquish control or was largely

absent, teacher leadership was somewhat stifled and minimized. Fundamentally,

then, teacher-leader-driven change must consist of a coalition of teacher leaders

with the principal playing nothing more nor less than a supportive role in those

particular change efforts. One reason why it is important to have the most powerful

player in the organization, such as the principal, on-board with the change is that

having the principal on board symbolically adds value to the change and makes the

improvement efforts more likely to take broad effect. We saw this at Dogwood

where the principal’s support for the PLC work was ongoing, and yet she made way

J Educ Change (2016) 17:85–113 105

123

for the PLC leader to develop and act on a vision for change. By contrast, the

principals’ roles within the guiding coalitions at the other two schools did not

support teacher leadership in promoting discussion-based teaching. At Maple, the

principal positioned herself as the director of her teacher leaders’ work, rather than

supporting it from a distance and offering support. At Spruce, given that the

principal was largely absent, the opposite was true. There was plenty of space for

teacher leadership but few structures or gestures of support and guidance.

Collectively then, these cases suggest the critical importance of the school

principal’s orientation toward leadership in creating and maintaining conditions that

enable authentic teacher leadership.

The leadership team as a system

Beyond just the principal’s orientation toward leadership, the dynamics among the

full guiding coalition—the principal and the teacher leaders—played a key role in

whether or not the leadership team worked harmoniously to propel change.

Figures 2, 3, and 4 convey our sense of how each leadership team functioned. We

found that the team that had the greatest success in implementing discussion-based

teaching was at Dogwood (Fig. 2), where there was a strong sense of momentum

toward discussion-based teaching and a high level of cohesion among four of the

five team members as they consistently reinforced the same vision of effective

teaching. The team also had a clear champion for the work in their PLC leader, and

the principal struck a balanced position of providing guidance and support while

enabling autonomy. Although individual team members expressed strong commit-

ments toward the change endeavors at the other two schools, the team dynamics at

Maple and Spruce (Figs. 3, 4) were much less cohesive in how the members

operated as a unit and less focused in their pursuits, both of which undermined their

ability to develop and communicate a consistent vision. As such, we found that only

Dogwood actually had a clear vision that could be captured in a few, succinct words

and clearly communicated to staff. Notably, Kotter (1996) proposes that formulating

and then communicating a vision for change are pivotal. If this is indeed the case,

then it is not surprising that broad instructional change did not emerge at either

Spruce or Maple, where we found little evidence of cohesive guiding coalitions or

coherent visions of effective instruction.

The school as a system

We found that the teacher leadership process was greatly shaped by and dependent

upon the contextual conditions within the schools, as prior research has noted, but

we also found that those conditions were not stable or transparent. The rapidly

changing context at Spruce Academy illustrates how quickly the foundational

conditions for teacher leadership described by York-Barr and Duke (2004) can

change and how such rapid change can undermine the potential impact of teacher

leadership. Considering these foundational conditions as ‘‘pre-requisites’’ for

effective teacher leadership, as we did in identifying three schools to study, turned

out to be inaccurate. Conditions within schools are not established and solidified

106 J Educ Change (2016) 17:85–113

123

before teacher leadership begins, but are constantly shifting. As Payne (2008) and

Ingersoll (2001) note, urban, high-poverty schools in particular have rapidly

changing contexts due to the high volume of turnover among students, teachers, and

administrators. Given this lack of stability, teacher leadership does not occur in the

linear fashion suggested by York-Barr and Duke’s graphic depiction of their model,

in which arrows pointing to the right convey a step-by-step process. Instead, we

found that the volatility of life in urban schools makes complex systems theory

(Opfer and Pedder 2011) a more useful model for understanding the dynamic role of

context in teacher leadership because it allows us to account for the multitude of

challenges that can impede positive change.

In addition to finding that foundational conditions can change rapidly, we also

found that such conditions were difficult to identify from the outside, even when

assessing a more stable school environment like Maple Academy. As PD providers

working with Maple for over a year, we invited the leadership team to participate in

our study because it appeared they met the foundational conditions in school

culture, relationships, and structures (York-Barr and Duke 2004). Ultimately,

however, we found that leadership at Maple was not shared among the principal and

teacher leaders; rather, it was primarily the domain of the principal. From the

outside, this hierarchical leadership was not apparent to us prior to the study, and we

question the extent to which it was even acknowledged and voiced among members

of the leadership team. If teacher leaders themselves do not realize that the nature of

leadership in their school is not well suited for teacher leadership, their efforts to

lead may be ineffective, regardless of how they enact the process. In such a case,

even attempting to enact Kotter’s (1996) strategies for leading change could have

little impact if the setting is not conducive to teacher-leadership-driven change.

Similarly, if teacher leaders do see limitations to their leadership potential due to the

principal’s style, the power imbalance between the teacher leader and the principal

can make this something that goes unspoken. Such power imbalances and an

associated lack of teacher involvement in decision-making are particularly common

in low-performing urban schools (Payne 2008), making such schools potentially

challenging environments for authentic teacher leadership. In such cases, teacher

leaders may take on leadership roles without any expectation for actually leading

change and may simply go through the motions. As such, inadequate school

conditions for effective teacher leadership are not necessarily observable or

acknowledgeable, despite the fact that these inadequacies can undermine teacher

leadership efforts.

The external context as a system

Kotter (1996) asserts that the first step in leading change is to establish a sense of

urgency that makes members of the organization strongly believe in a pressing need

for change. Across our three school cases, we believe this was done most effectively

at Dogwood, where Daphne positioned the vision of Bridging the System as being

directly in response to students’ learning needs. This localization of urgency

appeared to create a compelling rationale for teachers to wholeheartedly engage in

professional activities that would serve the school’s particular student population.

J Educ Change (2016) 17:85–113 107

123

By contrast, we found that urgency at Maple centered on implementing the

principal’s many initiatives, and urgency at Spruce Academy was around

acclimating new teachers. Certainly, teacher leaders in these schools pursued these

objectives in purposeful ways with the ultimate goal of improving instruction for

students. Yet, Kotter’s theory suggests that the Maple and Spruce leadership teams

made a misstep by not creating a more compelling sense of urgency that focused on

the external context of their schools. Like Dogwood, Maple and Spruce are urban

charter schools serving high-needs populations with low test scores—particularly in

math, where none of the schools had more than 15 % of students scoring advanced

or proficient on state assessments. Yet, the foci of urgency at these two schools

seemed decontextualized from this reality. We believe that the contexts at Maple

and Spruce could have been used to make a case for increasing the quantity and

quality of discussion-based teaching, but that these opportunities were missed. At

Maple, although teachers experienced pressure to meet the principal’s expectations

of raising student test scores, there was no contextual rationale for mastering any

particular teaching practice to serve students’ needs. At Spruce, helping new

teachers develop their instructional skills in one particular area—such as discussion-

based teaching—could have increased their success in the classroom while

simultaneously helping acclimate them to the profession through collectively

developing shared instructional skills. In pointing out these missed opportunities,

we do not mean to discredit the important goals of promoting good teaching

practices and building staff morale and stability, which we applaud Maple and

Spruce for taking on. Rather, we suggest that utilizing the contexts of their schools

to provide rationale for these goals could have propelled these efforts even further.

In considering whether and how the external context impacted how the leadership

teams framed their senses of urgency, we could not help but note that the only

school that focused on the specific needs of their mostly African American students

was the one school staffed by primarily African American women. In the other two

schools, where the faculty was primarily white, we noted no references to the

specific needs of ‘‘our students’’ in the videos, reflections, or interviews. Although

we hesitate to draw conclusions based on omissions that may simply be a product of

our data collection procedures, we suggest that future research examine this

potentially compelling link between how teacher leaders identify with the local

community and how they frame the urgency for instructional change. It may be the

case that some of the impediments to urban school reform described by Payne

(2008)—including low expectations for students or pessimistic views on reforms—

could vary depending on the demographic composition of the educators. Research

examining this possibility might help us better understand the role of the external

context in teacher leadership among teachers from various racial and ethnic

backgrounds who are working in urban communities.

Teacher leadership for influence or change?

In conducting this study, we sought to examine the assumptions that teacher leaders

both have a means to influence their colleagues’ work and engage in actions that

lead their colleagues to change their practice. Given these assumptions, we looked

108 J Educ Change (2016) 17:85–113

123

for both means of influence, which we conceptualized as opportunities to share

one’s practice and inform the thinking of others, and actions that lead to change,

which we identified as specific tactics that propel others to do something different in

a specific, intentioned way. During conversations among the members of our

research team, we discussed whether or not influence and change are substantively

different concepts and whether the optimal goal of teacher leadership is to influence

or change the instruction of the teacher leaders’ colleagues. Importantly, York-Barr

and Duke’s model examines teacher leadership influence, whereas Kotter’s (1996)

eight steps target change. As we considered our data using these varying lenses, we

debated whether the differences between the two terms were more than simply

semantics, and we found it useful to consult the definitions provided by Merriam-

Webster’s online dictionary. This source defines influence as ‘‘to affect or change

someone or something in an indirect but usually important way’’ and change as ‘‘to

make someone or something different.’’ We see the distinction between these two

words as being the indirect nature of influence and the more intentional, direct

nature of change. Given the intentional nature of change, we inferred that teacher

leaders engaged in conscious change efforts were seeking to foster more substantial,

specific changes in teaching practice. They had a specific end goal in mind—a

vision for change—that transcended simply wanting their colleagues to learn some

new techniques and practices.

Our comparisons of teacher leadership across the three schools furthered this

influence versus change distinction. We found Daphne’s Bridging the System

approach at Dogwood to be most strongly aligned with Kotter’s (1996) first four

steps for leading change, and we felt convinced that broad-based change toward

discussion-based teaching was well underway. As evidence, our videos captured

Dogwood teachers examining the nuances of discussion-based teaching, such as

during a PLC conversation in response to the question ‘‘What are some possible

reasons why a student might not feel comfortable talking in class?’’ and in a lesson

debrief in which a mentor and mentee discussed why a basketball review game

might not be the best way to get students to collaborate in preparation for a social

studies test. Throughout such interactions, it was clear that Dogwood teachers were

integrating discussion-based teaching into their instruction and grappling collec-

tively with some of the challenges of changing one’s practice in these specific ways.

By contrast, the teacher leadership efforts at Spruce and Maple appeared directed

toward the less intentional, less impactful outcome of influence. This was certainly

the case at Spruce, where given Stacy’s passive language and presentation of

teaching ‘‘tips,’’ her primary means of changing teaching would be through indirect

influence if someone were to try a strategy she shared and ultimately integrate it into

their practice. At Maple, the emphasis on influence rather than change was evident

in the disconnected nature of multiple initiatives that confused the vision and

overextended teachers’ abilities to focus on particular changes in their teaching.

Whereas teacher leaders at Spruce seemed to purposefully limit their efforts to

influence, the team at Maple seemed to default to influence due to an ineffective

approach to producing change, even when direct change in teaching practice

appeared to be their objective.

J Educ Change (2016) 17:85–113 109

123

We posit that actors targeting either influence or change might have different

expectations for the depth and gravity of the outcomes of teacher leadership.

Influence suggests that teachers ultimately integrate some new practices into their

teaching—akin to Piaget’s (2000) concept of assimilation of new knowledge that

becomes incorporated into existing schema within the learner’s mind. Yet, change

implies a more emboldened reframing of a teacher’s instruction—akin to Piaget’s

concept of accommodation, wherein the learner reconfigures their mental schema to

represent altered understanding of a concept. We assert that this influence versus

change distinction could be critical to understanding the teacher leadership process

if intentional, direct change requires decisive, clear action, as our study suggests. In

this way, York-Barr and Duke’s (2004) language of maintaining a focus on teaching

and learning, establishing trusting and constructive relationships, and interacting

through formal and informal points of influence really does seem to target influence.

This is not surprising, as the model uses the term influence to denote the goal of

teacher leadership. But, we question whether influence is too passive of a concept to

generate real change, and given the great need for instructional improvement in

urban schools, we propose that the more intentional, direct nature of targeted

change in instruction might be the more compelling objective for teacher

leadership.

Conclusion

Our findings on the teacher leadership processes within these three urban charter

schools suggest that broad-based instructional change requires teacher leaders to be

purposeful and focused in creating change through targeted, direct, and strategic

change efforts. To this end, we found that Kotter’s (1996) eight steps for generating

organizational change offered insight into the process by which teacher leaders can

undertake instructional change efforts, and we found complex systems theory

(Opfer and Pedder 2011) to be an informative way to conceptualize the influence of

multiple embedded systems on the teacher leadership process. By integrating these

theoretical lenses, our work complicates and provides insight into the means by

which teacher leadership creates change. Just the same, we acknowledge that in the

complex contexts facing urban schools with low student achievement and high

teacher turnover, understanding the teacher leadership process for leading change

will ultimately require more than the theories of Kotter (1996) and Opfer and Pedder

(2011) can provide. Urban, high-needs contexts continue to be some of the most

challenging, yet critical, places for enacting educational change. Although we have

sought to explain elements of the teacher leadership process in these settings, there

is still more work to be done to understand how school and teacher leaders can

operationalize positive instructional change in urban schools.

In our own PD work, however, we have found direct applications of this research

that have helped us improve the support and preparation we provide formal teacher

leaders. In subsequent PD sessions, we have paid greater attention to not only

providing extensive information on the rationale for discussion-based teaching, but

also to assessing that understanding among our participants and providing

110 J Educ Change (2016) 17:85–113

123

opportunities for participants to engage in metacognition about their own

understanding and learning regarding discussion-based teaching. We have also

provided teacher leaders with information on leading change and have specifically

introduced them to Kotter’s (1996) eight steps for leading organizational change. In

coaching teacher leaders in the final 2 years of our PD, we purposefully integrated

Kotter’s terminology—urgency, guiding coalition, vision, etc.—in an effort to

increase the impact of teacher leaders in their schools and to help them navigate the

complex systems that frame their work. We believe that this more strategic focus on

the preparation of teacher leaders to fully and effectively engage in the change

process is a critical future direction for research and practice in teacher leadership.

Acknowledgments The authors gratefully acknowledge funding from the Spencer Foundation, which

made this research possible.

References

Angelle, P. S., & Schmid, J. B. (2007). School structure and the identity of teacher leaders: Perspectives

of principals and teachers. Journal of School Leadership, 17(6), 771–799.

Ball, D. L., Sleep, L., Boerst, T., & Bass, H. (2009). Combining the development of practice and the

practice of development in teacher education. Elementary School Journal, 109(5), 458–474.

Beachum, F., & Dentith, A. (2004). Teacher leaders creating cultures of school renewal transformation.

The Educational Forum, 68, 276–286.

Borko, H., Jacobs, J., Eiteljorg, E., & Pittman, M. (2008). Video as a tool for fostering productive

discussions in mathematics professional development. Teaching and Teacher Education, 24,

417–436.

Chenoweth, C. (2009). How it’s being done: Urgent lessons from unexpected schools. Cambridge, MA:

Harvard Education Press.

Cranston, N. C. (2000). Teachers as leaders: A critical agenda for the new millennium. Asia-Pacific

Journal of Teacher Education, 28(2), 123–129.

Crowther, F., Kaagan, S. S., Ferguson, M., & Hann, L. (2002). Developing teacher leaders. Thousand

Oaks, CA: Corwin Press.

Darling-Hammond, L., Wei, R. C., Andree, A., Richardson, N., & Orphanos, S. (2009). Professional

learning in the learning profession: A status report on teacher development in the US and abroad.

NSDC. Retrieved from National Staff Development Council website: http://www.nsdc.org/news/

NSDCstudytechnicalreport2009.pdf.

Dewey, J. (1933). How we think. Buffalo, NY: Prometheus Books. Original work published 1910.

Drago-Severson, E. (2007). Helping teachers learn: Principals as professional development leaders.

Teachers College Record, 109(1), 70–125.

Fairman, J. C., & Mackenzie, S. V. (2012). Spheres of teacher leadership action for learning. Professional

Development in Education, 38(2), 229–246.

Fairman, J. C., & Mackenzie, S. V. (2014). How teacher leaders influence others and understand their

leadership. International Journal of Leadership in Education,. doi:10.1080/13603124.2014.904002.

Frost, D., Durrant, J., Head, M., & Holden, G. (2000). Teacher led school improvement. New York:

Falmer.

Gambrell, L. B., & Almasi, J. F. (Eds.). (1996). Lively discussions! Fostering engaged reading. Newark,

DE: International Reading Association.

Grissom, J. A., & Loeb, S. (2011). Triangulating principal effectiveness: How perspectives of parents,

teachers, and assistant principals identify the central importance of managerial skills. American

Educational Research Journal, 48(5), 1091–1123.

Grossman, P., Hammerness, K., & McDonald, M. (2009). Redefining teaching, reimagining teacher

education. Teachers and Teaching, 15(2), 273–289.

J Educ Change (2016) 17:85–113 111

123

Heck, R., & Hallinger, P. (2009). Assessing the contribution of distributed leadership to school

improvement and growth in math achievement. American Educational Research Journal, 46(3),

659–689.

Horn, I., & Little, J. (2010). Attending to problems of practice: Routines and resources for professional

learning in teachers’ workplace interactions. American Educational Research Journal, 47(1),

181–217.

Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analysis. American

Educational Research Journal, 38(3), 499–534.

Jadallah, M., Anderson, R. C., Nguyen-Jahiel, K., Miller, B. W., Kim, I., Kuo, L., et al. (2010). Influence

of a teacher’s scaffolding moves during child led small group discussions. American Educational

Research Journal, 48(1), 194–230.

Kardos, S. M., Johnson, S. M., Peske, H. G., Kauffman, D., & Liu, E. (2001). Counting on colleagues:

New teachers encounter the professional cultures of their schools. Educational Administration

Quarterly, 37(2), 250–290.

Katzenmeyer, M., & Moller, G. (2001). Awakening the sleeping giant: Helping teachers develop as

leaders. Newbury Park, CA: Corwin Press.

Kotter, J. P. (1996). Leading change. Cambridge, MA: Harvard Business Review Press.

Lampert, M., Boerst, T., & Graziani, F. (2011). Organizational resources in the service of school-wide

ambitious teaching practice. Teachers College Record, 113(7), 1361–1400.

Lieberman, A., & Friedrich, L. D. (2010). How teachers become leaders: Learning from practice and

research. New York: Teachers College Press.

Little, J. W. (2006). Professional community and professional development in the learning-centered

school. National Education Association. Retrieved from www.nea.org.

Mangin, M. M. (2007). Facilitating elementary principals’ support for instructional teacher leadership.

Educational Administration Quarterly, 43(3), 319–357.

Margolis, J., & Huggins, K. S. (2012). Distributed but undefined: New teacher leader roles to change

schools. Journal of School Leadership, 22(5), 953–981.

Matsumura, L., Garnier, H., Correnti, R., Junker, B., & Bickel, D. (2010). Investigating the effectiveness

of a comprehensive literacy coaching program in schools with high teacher mobility. The

Elementary School Journal, 111(1), 35–62.

Maxwell, J. A. (2005). Qualitative research design: An interactive approach. Thousand Oaks, CA: Sage.

Miles, M., & Huberman, A. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.).

Thousand Oaks, CA: Sage.

Murphy, J. (2005). Connecting teacher leadership and school improvement. Thousand Oaks, CA: Corwin

Press.

Opfer, V. D., & Pedder, D. (2011). Conceptualizing teacher professional learning. Review of Educational

Research, 81(3), 376–407.

Paine, L., Pimm, D., Britton, E., Raizen, S., & Wilson, S. (2003). Rethinking induction: Examples from

around the world. In M. Scherer (Ed.), Keeping good teachers (pp. 67–80). Washington: ASCD.

Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). Thousand Oaks, CA: Sage.

Payne, C. M. (2008). So much reform, so little change: The persistence of failure in urban schools.

Cambridge, MA: Harvard Education Press.

Piaget, J. (2000). Piaget’s theory of cognitive development. In K. Lee (Ed.), Childhood cognitive

development: The essential readings (pp. 33–47). Malden, MA: Blackwell.

Stanulis, R. N. & Brondyk, S. (2013). Complexities involved in mentoring towards a high leverage

practice in the induction years. Teachers College Record, 115(10), http://www.tcrecord.org. ID

number: 17157.

Stanulis, R. N., Brondyk, S., Little, S., & Wibbens, E. (2014). Mentoring beginning teachers to enact

discussion-based teaching. Mentoring and Tutoring: Partnership in Learning, 22, 127–145.

Stanulis, R. N., Meloche, B., Robinson, S., & Prybys, J. (2011). Rubric for guiding principal selection of

teacher leaders. Detroit, MI: Presentation at MAPSA Principal Institute.

Stein, M. K., & Nelson, B. S. (2003). Leadership content knowledge. Educational Evaluation and Policy

Analysis, 25(4), 423–448.

Timperley, H. S. (2005). Instructional leadership challenges: The case of using student achievement

information for instructional improvement. Leadership and Policy in Schools, 4(1), 3–22.

Valli, L., van Zee, E., Rennert-Ariev, P., Mikeska, J., Catlett-Muhanunad, S., & Roy, P. (2006). Initiating

and sustaining a culture of inquiry in a teacher leadership program. Teacher Education Quarterly,

33(3), 97–114.

112 J Educ Change (2016) 17:85–113

123

Wood, D. (2007). Teachers’ learning communities: Catalyst for change or a new infrastructure for the

status quo? Teachers College Record, 109(3), 699–739.

Yin, R. K. (2009). Case study research: Design and methods (4th ed., Vol. 5). Thousand Oaks, CA: Sage.

York-Barr, J., & Duke, K. (2004). What do we know about teacher leadership? Findings from two

decades of scholarship. Review of Educational Research, 74(3), 255–316.

Yost, D., Vogel, R., & Rosenberg, M. (2009). Transitioning from teacher to instructional leader. Middle

School Journal, 40(3), 20–27.

J Educ Change (2016) 17:85–113 113

123

Copyright of Journal of Educational Change is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.