Ending Isolation:

The Payoff of Teacher Teams in Successful High-Poverty Urban Schools

Susan Moore Johnson

Stefanie K. Reinhorn

Nicole S. Simon

Working Paper

The Project on the Next Generation of Teachers

Harvard Graduate School of Education

July 2015

Abstract

This qualitative analysis of teacher teams is part of a larger, comparative case study, Developing Human Capital Within Schools, conducted by the Project on the Next Generation of Teachers. Within one city, we interviewed 142 teachers and administrators in six high-poverty schools (traditional, charter, and turnaround), all of which had achieved the highest ranking in the states accountability system. Here, we analyze the role that teacher teams played in the experiences of teachers. Although all six schools emphasized the value of collaboration among teachers, only five relied on teams as a means for school improvement, dedicating substantial, regular blocks of time for teachers to meet. Teams focused on matters of content (curriculum, lesson plans, and student achievement) and the student cohort (individual progress, group behavior, and organizational culture). Teachers valued their work on teams, saying that it supported their instruction and contributed to the schools improvement by creating coherence across classrooms and shared responsibility for all students. Factors that supported teams included having a worthy purpose in support of the schools mission; sufficient, regular time for meetings; engaged support by administrators; facilitation by trained teacher leaders; and integration of the teams work with other supports for teachers. This working paper is part of a larger study conducted by researchers at the Project on the Next Generation of Teachers. We are indebted to the Spencer Foundation and to the Harvard Graduate School of Education for funding this project, although all views presented here are our own. We appreciate the comments and suggestions of Judith Warren Little at the Annual Meeting of the American Educational Research Association, 2015. We are deeply grateful to the administrators and teachers who participated in this study.

Ending Isolation

2

Introduction

Instructional teams for teachers have emerged as a central component in many schools

strategy for improvement. Grounded in evidence about the benefits of ongoing collaboration

(Bryk, Sebring, Allensworth, Luppescu, & Easton, J.Q, 2010; Rosenholtz, 1989; Little, 1982),

teams are intended to decrease the professional isolation of teachers (Lortie, 1975) and

substantially reduce the well-documented variation in teachers effectiveness across classrooms

(Rivkin, S. G., Hanushek, E. A., & Kain, J. F., 2005; Rockoff, J.E., 2004). Given pressing

demands for accountability and the far-reaching needs of urban students who live in poverty,

teachers instructional teams may prove to be a valuable mechanism for improving urban

education, increasing the likelihood that all students in all classrooms will be well served. As

they work together to implement new programs and policies, teachers with different areas of

expertise and levels of experience may find that teams not only provide guidance about

curriculum development, lesson planning, and pedagogy, but also are a source of professional

and personal support that sustains them in their work and improves the professional culture in

their school.

Despite such promise, most schools are not currently organized so that teachers

collaborate regularly and intensively. In fact, some research suggests that it may be unrealistic to

expect teachers to engage with teams that require more than sharing materials or advice (Little,

1990) because interdependence runs counter to professional norms of autonomy and privacy,

which have long defined teachers work (Donaldson, Johnson, Kirkpatrick, Marinell, Steele, and

Szczesiul, 2008; Little, 1990; Lortie, 1975;). When teachers participate on teams, they often tend

to keep their colleagues at arms length, protecting the core of their work from interference. They

may discuss curricular options or suggest ideas and techniques that seem promising (Charner-

Ending Isolation

3

Laird, Ng, Johnson, Kraft, Papay, & Reinhorn, 2015; Troen & Boles, 2012). Yet they seldom

relinquish their instructional autonomy, ultimately returning to their classroom to teach as they

know best. Most teachers tacitly understand, accept, and enact this approach to participating with

colleagues.

However, the role of teacher teams appears to be rapidly evolving in many schools that

serve high-need students. In a prior study that we conducted about teachers experiences in six

high-poverty schools of one urban district (Charner-Laird et al., 2015), all teachers reported

having time allocated weekly for team meetings. Overall, the teachers interviewed for that study

used two criteria to assess their teams goodness of fit with the demands of their work: (1) Did

their team help them improve their own teaching? (2) Did their teams work contribute to

improving their school? Although all the teachers interviewed for that study saw great potential

in teams, those in only three schools reported that their teams were effective in helping them

improve their daily instruction while also contributing to the schools progress in meeting its

goals. Teachers in the other three schools said their teams were ineffective, either because their

principal had defined no meaningful purpose for their work together, excessively prescribed and

monitored their teams process, or neglected to support them if they took risks to improve

instruction.

Our current study is broadly focused on the human capital practices in six different

schools (including traditional, turnaround, and charter schools) all of which served large

proportions of students living in high-poverty, high-minority communities. However, each

school in the current study had demonstrated success with its students by achieving the highest

performance rating in the states accountability system. We sought to understand that success,

specifically what these schools did to recruit, hire, support, and develop their teachers. This

Ending Isolation

4

analysis centers on the use of teacher teams to improve teaching and learning. Were teams

central to these schools strategy for improvement? If so, what role did they play and how did

teachers experience and assess them?

We found that all six schools valued and encouraged collaboration among teachers, but

that only five regularly scheduled and purposefully used teams as a prominent mechanism for

improvement. The sixth school promoted collaboration in less formal ways. Teams in these five

schools had two areas of focus. The first was instructional content, which included curriculum

development, lesson planning, and regular review of data about students learning. The second

was the student cohort, which included individual students well-being and progress, the cohorts

behavior and compliance with rules, and the organizational culture that students in the cohort

experienced. Teachers in these five schools offered very positive assessments of their work

together, suggesting that their teams had achieved a goodness of fit (Charner-Laird et al.,

2015) in meeting their individual instructional needs and addressing the schools needs for

improvement.

In what follows, we first review the background and context of this study, considering

evidence about the potential of teacher collaboration for improving schools and about the

challenges facing those who rely on teams as a centerpiece of their work. Although the track

record for teacher teams has been generally weak and disappointing, we suggest that current

changes in the conditions of schooling may improve the prospects for teams success. After

describing our research methods, we introduce the schools of our study and present our findings.

We explain why teachers judged their teams to be effective in addressing matters of content and

the student cohort. In doing so, we note important differences in how schools organized their

teams, given the time and flexibility available to them. After discussing the factors that we

Ending Isolation

5

found contributed to a schools effective reliance on teams, we conclude by considering the

implications of this research for policy, practice, and research.

Background and Context

Beneath the question about whether teachers successfully engage in teams lies the more

fundamental issue of whether teachers actually want to collaborate with colleagues. For if

teachers resist collaboration, either tacitly or outright, it may be futile for schools to invest scarce

resources in trying to establish teams. If, however, teachers see potential in teams, but are

deterred from seriously engaging with colleagues because of the limitations within their school

scarce time, overwhelming demands, competing purposes, or lack of effective leadershipthen

it is important to better understand what it might take to address those factors so that teams could

reach their promise as a strategy for school improvement.

Evidence about teachers attitudes toward collaboration is mixed. Teachers repeatedly

report that they depend on colleagues and value working with them (Drury & Baer, 2013; Kardos,

Johnson, Peske, Kauffman, & Liu, 2001). Yet 25 years after Littles (1990) clear-eyed analysis

of teachers collaborative practices, examples of serious, sustained collaboration among teachers

remain relatively few. Having explored the interplay between teachers professional norms and

their workplace, Little concluded in 1990: Schoolteaching has endured largely as an

assemblage of entrepreneurial individuals whose autonomy is grounded in norms of privacy and

noninterference and is sustained by the very organization of teaching work (p. 550). Few would

argue that the situation changed markedly during the decade following Littles observation.

However, as we assess the prospects for teacher teams today, we do so in a different

context. Since 2000, the composition of the teaching force has changed rapidly with the entry of

a different generation of teachers (Johnson & The Project on the Next Generation of Teachers,

Ending Isolation

6

2004). At the same time, state and federal accountability policies have set unprecedented

expectations for assessing both students learning and teachers performance. Also, in response

to charter schools and federal efforts to improve failing schools, additional resources and more

autonomy have become available to some schools, freeing up constraints on staffing, time, and

budget allocations that, in the past, interfered with teams development. These changes lead us to

ask whether teams for teachers have the potential for greater success today than in the past.

The Structure of Schooling and Organization of Teaching

Tyack (1974) explains that as the burgeoning US population moved to the cities in the

late 19th century, the one-room schoolhouse served as the building block for age-graded schools.

Assigning teachers and their students to separate classrooms that ran along both sides of a

corridor created the so-called egg-crate school (p. 44), a physical and organizational structure

that was efficient, but proved to be an isolating, inhospitable environment for collaboration.

Many teachers hired in the late 1960s and early 1970sthe cohort now retiringquickly learned

that a career in teaching required them to function independently or, at best, with the approval of

the principal and the ongoing encouragement of one or two peers. In 1990, Little reported that

the modal reality of teachers work was independence punctuated by occasional contacts

among colleagues (p. 513). Not only has the compartmentalized structure of schooling allowed

teachers to operate independently, it has virtually obliged them to do so (Johnson, 2012; Lortie,

1975).

Collaboration among teachers. Deeply rooted norms of autonomy and privacy among

teachers were well established in the schools of the 1970s and have continued to be a fearsome

deterrent to collaboration. Researchers widely agree that, despite enthusiasm about the benefits

of collaboration, few teachers participate in what Little (1990) calls joint work (p. 520). Some

Ending Isolation

7

scholars, notably Huberman (1993), argue that because teachers are essentially artisans who

develop a personal craft, deliberate efforts to coordinate their instructional practice are

unwarranted. However, others have documented the positive role that collaboration can play in

school improvement (Little, 1982; Rosenholtz, 1989; Newmann & Wehlage, 1995; McLaughlin

& Talbert, 2001; Bryk, et al., 2010). For example, Rosenholtz (1989) details how a shared sense

of purpose among teachers in the high consensus schools she studied led to collaborative

school cultures that were learning enriched (p. 206), while little common purpose existed in

low consensus schools and few teachers seemed attached to anything or anybody.

Rosenholtz characterized these failing schools as learning impoverished for teachers and

students alike (p. 207).

In 1995 Newmann and Wehlage advanced our understanding of how school context

supports students learning with their longitudinal case studies of 32 schools, all involved in

restructuring. Based on their findings from these schools, supplemented by surveys from the

Chicago Study of School Reform and data from National Educational Longitudinal Study of

1988, these researchers reaffirmed the importance of teachers shared responsibility for school

improvement: [T]eachers in these schools take collectivenot just individualresponsibility

for student learning, and for constantly improving their teaching practice (p. 4). Newmann and

Wehlage conclude that schools benefit from having interdependent work structures, such as

teaching teams, which encourage collaboration (p. 2).

Recently, Bryk and colleagues (2010) reported further evidence of the importance of

collaborative work among teachers and administrators with their analysis of seven years of

longitudinal data on Chicago schools practices and student achievement. They conclude that

students benefit when teachers collaborate about curriculum, instruction, and problems of

Ending Isolation

8

practice within school contexts that are grounded in strong norms of trust, respect and continuous

improvement.

Other scholars have reported a positive relationship between collaboration among

teachers and their students achievement (Gallimore, Ermeling, Saunders, & Goldenberg, 2009;

Goddard, Goddard, Tschannen-Moran, 2007; Parise & Spillane, 2010). Jackson and

Bruegmanns (2009) large-scale analysis of statewide data from North Carolina revealed that,

when a more effective teacher joins a schools grade level, all classes of students at that grade

level make larger achievement gains in English language arts and mathematics, both initially and

over time. The authors call these widespread, positive effects, peer-induced learning (p. 87).

Their study, however, provides no means to explain the policies, structures, or leadership that

generate such benefits.

Changing Conditions May Improve Prospects for Teacher Teams

Since 2000, changes in the context of schooling and the work of teachers may have

improved the prospects for teams success. We consider here changes in the composition of the

teaching force; state and federal demands for accountability that arguably cannot be met by

teachers working in isolation; and new management autonomies granted to some schools as they

work to improve.

The expectations of a new generation of teachers. One important question is whether

teachers who entered the profession since 2000 have different expectations for their career and

their workplace than those of the veterans they replaced. In part, this is a question about whether

the norms of autonomy and privacy in teaching are immutable, or whether broader changes in

social conventions, teacher labor markets, and education policy moderate or redirect those long-

standing expectations and norms.

Ending Isolation

9

In 1998, we conducted a four-year longitudinal study of 50 first- and second-year

teachers in Massachusetts (Johnson & Project on the Next Generation of Teachers, 2004). These

novices reasons for entering teaching were much like those of veteran teachers; they wanted to

work with youth, to contribute to the social good, and to share a love of their subject (Johnson,

1990). However, the new teachers had different expectations for their school, they described

different responses to their early years in the classroom, and they ultimately made different

choices about their career than their predecessors. Although the veterans they replaced had

decided to teach when few professional options were available to women and to men of color,

recent novices chose teaching from among many alternatives. They explained that, if they were

dissatisfied with teaching and found it impossible to achieve a sense of success in the

classroom (Johnson & Birkeland, 2003), they would leave teaching. Serial careers were

increasingly common for many in their generation. Even if they remained in teaching, these

teachers predicted that they would not stay in the classroom full-time, anticipating instead roles

for themselves as instructional coaches or teacher leaders, which would allow them to vary their

work and extend their influence beyond their classroom. As youth they had played on sports

teams and as college students they had produced team-based projects. Given such experiences,

new teachers expected to work closely with their colleagues and, in fact, dreaded the prospect of

being isolated in their classroom (Kardos & Johnson, 2007). When we concluded the study four

years later, 17 of the original 50 teachers remained in their initial school, 16 had changed

schools, and 17 had left public school teaching (Johnson & Birkeland 2003). These proportions

were roughly comparable to those of teachers nationally (Leukens, Lyter, Fox, & Chandler,

2004). The primary explanation teachers offered for leaving their school or teaching was that

Ending Isolation

10

their workplace failed to support them in effectively doing the work they had chosen (Johnson &

Project on the Next Generation of Teachers, 2004; Johnson & Birkeland, 2003).

Some from this new generation assumed roles as teacher leaders during the second stage

of their career, years 4-10 (Donaldson et al., 2008). However, they often encountered subtle or

overt resistance from the colleagues whose practice they were supposed to influence in their new

role. Although such challenges may have slowed the progress of teacher leadership, they did

reveal the tension within many schools between the retiring generations norms of privacy and

autonomy and early-career teachers expectations for collegiality and interdependence. As time

passes, increasing numbers of new teachers with an interest in collaboration might open the way

for greater interdependence among teachers in their work. In fact, recent research suggests that

such change is underway in some settings (Grossman, P., Wineburg, S., & Woolworth, S, 2001;

McLaughlin, & Talbert, 2006; Horn & Little 2010; Coburn, C.E., Mata, W.S., & Choi, 2013).

The press of public policy. In 1990, Little conjectured that interdependent work among

teachers might increase in response to external demands for improved performance: As

policymakers and the wider public press schools to achieve more ambitious and complex goals,

school leaders in turn press teachers to collaborate in the service of those goals (p. 320). Such

press built steadily, beginning with publication of A Nation at Risk in 1983 and continuing

through state accountability policies of the late 1990s, the federal No Child Left Behind Act of

2001(NCLB), and the Race to the Top competition in 2010. Increasingly students standardized

test scores have been used to assess schools, with those that failed to make adequate yearly

progress under NCLB being warned or sanctioned. Before such policies, teachers might safely

do their best in their own classroom, without feeling obliged to confer with colleagues about

their shared responsibility for all the schools students. Over time, however, federal, state, and

Ending Isolation

11

local policies have increasingly influenced teachers practices. When Cohen and Hill (2001)

studied the influence of policy on practice in Californias program to improve mathematics

teaching, they found that the policy was most likely to be implemented successfully when it was

accompanied by extended opportunities for professional learning that [were] grounded in

practice (p. 10). Recently, Coburn, Mata, & Choi (2012) tracked the effect of district policy on

teachers professional networks and found that teachers changed their patterns of interaction and

were more amenable to outside influence than previously thought. Social policy, the authors

concluded, can play a role in fostering conditions in schools within which teachers seek out

their colleagues, share information, solve problems, and learn from one another in their

networks (p. 331).

Organizational structures in schools, including teaching schedules and course

assignments, have been partially responsible for low rates of collaboration among teachers.

Certainly, one of the most problematic features of teachers work is the rushed schedule and

shortage of predictable time for them to work together. International comparisons reveal that

U.S teachers have far less time available for, and committed to, collegial work than their

counterparts abroad (Organisation for Economic Cooperation and Development, 2005; Darling-

Hammond, Wei, Andree, Richardson, & Orphanos, 2009). Given the scarcity of formal

preparation time during the school day, U.S. teachers typically communicate with colleagues at

lunch, on the playground, or fleetingly between classes. However, in response to accountability

policies such as NCLB, which require that schools achieve success with all sub-groups of

students, many schools have begun to schedule common planning time when teachers of the

same students, grades levels, or subjects can meet to learn from one another and coordinate their

work.

Ending Isolation

12

Recent polices provide some principals with increased resources and flexibility. The

formal authority that states grant charter schools to choose staff, budget resources, and schedule

time has had spillover effects in some conventional public schools. Many states and school

districts today authorize various schools to exercise a range of autonomies similar to those of

charters. For example, in 1994 the Boston Public Schools and the Boston Teachers Union

agreed to create a group of Pilot Schools that would be free from many restrictions imposed on

conventional district schools by the central office or the teachers contract (Pearlman, 2000). Pilot

Schools, which have the right to hire, fire and transfer teachers; budget funds; choose curriculum

and assessments; schedule time; and create their own a governing board, eventually appeared in

other urban school districts, including Los Angeles. State officials in Massachusetts, where our

study took place, not only authorized state-sponsored charter schools, but also established

procedures for creating Innovation Schools, which remain under the authority of the local school

district, but have charter-like flexibility. With such changes, the once distinct line between

charter schools and traditional district schools blurred. Drawing on the autonomies provided by

various state and local policies, many schools have established new models for collaboration

among teachers.

The federal government also contributed to reducing differences between charter and

traditional schools when it required applicant states to close or improve failing schools under the

Race to the Top competition and subsequent waivers. States and districts could designate

unsuccessful schools for turnaround (which required hiring a new principal and a staff that

included no more than half of the current teachers) or for restart (which required that the

school be managed by a new operator, often a charter management organization). In

Massachusetts, schools in turnaround competed for additional funding to support improvement

Ending Isolation

13

through federal School Improvement Grants, administered by the state. Principals also were

granted expanded autonomy to manage their school. For example, principals of turnaround and

restart schools gained the authority to rehire current teachers, hire new teachers, and dismiss or

transfer unwanted teachers. The three-year federal grants that these schools received to support

their redesign and improvement efforts often included funds to support additional professional

development for teachers or expanded learning time for students. Having more time and funding

for teacher teams brought unexpected benefits for failing schools that state officials said must

improve or be closed.

Combined, these changes in the context of teaching and schoolinga new generation of

teachers, increased demands for accountability, and additional resources and flexibility for both

charter schools and failing schoolshave increased the likelihood today that teachers will

accept, if not welcome, the opportunity to collaborate closely in teams with their colleagues.

Methods

This analysis draws on data collected for a larger study, Developing Human Capital

Within Schools, in which we broadly examined what six high-poverty, urban schoolsall of

which had received the states highest accountability ratingdid to attract, develop, and retain

teachers. Here we focus on these schools use of teams.

This analysis is guided by the following research questions:

1. Do these six successful schools rely on teams to develop teachers practice and increase the schools instructional capacity and coherence? If so, what purposes do the teams serve and how are they organized?

2. Are there notable school-to-school differences in how these teams are organized and managed? If so, what are they and what accounts for them?

3. How do teachers assess their experience with teams in their school? What role do

administrators play in teachers judgments about the value of teams?

Ending Isolation

14

4. Does policy play a role in the use or success of teams? Do charters, turnarounds,

and traditional schools approach instructional teams differently and, if so, in what ways and with what consequence?

The Sample Of Schools

Our sample selection was guided by four principles. First, we sought a sample that

included charter and district schools located in one city in Massachusetts. Second, we limited

our sample to schools that served high-poverty populations (where 70% or more of students are

eligible for free or reduced-price lunch) and communities of color. Third, we sought schools that

were successful, having achieved the highest rating in the states accountability system. Fourth,

we sought schools that employed distinctive approaches to human capital development.

To attend to the first three principles, we examined publicly available demographic and

student performance data. In seeking schools that were having success with students from

low-income families, we used the states accountability ratings as a proxy for students academic

achievement. The Massachusetts Department of Elementary and Secondary Education [MA

DESE] rates every school on a scale from 1 to 5, with 1 designating the highest performing

schools. The formula calculating a schools rating relies heavily on results from the

Massachusetts Comprehensive Assessment System [MCAS], the states high-stakes standardized

student test. The formula accounts for growth in student performance and the schools progress

in narrowing proficiency gaps among subgroups of students, using a weighted average from the

four most recent years of MCAS data. Although this definition of success relies heavily on

standardized test scores, which have limitations, it was the best proxy available for identifying

schools that have a positive impact on students academic outcomes. In addition, these were the

Ending Isolation

15

measures by which the schools were judged for awards and sanctions by the state and district, as

well as by funders, school boards and the media.

To attend to the fourth principle, we consulted our professional networks and considered

available information about the approaches to human capital development used by specific

schools and charter networks. Based on our initial inquiry, we drew up a proposed sample of six

schoolsall geographically located within the boundaries of one large urban school district,

Walker City School District [WCSD].1 The sample included three district schools (one

traditional and two former turnaround) and three state-authorized charter schools (including one

WCSD restart school). All were elementary and/or middle schools, which facilitated cross-site

comparisons. To recruit schools, we contacted school officials, explaining our study and

requesting their participation. All six schools that we approached agreed to participate. (For

descriptive statistics for sample schools see Table 1 below). The purposive nature of our sample

has allowed us to conduct an in-depth, exploratory study of schools in a particular context.

However, because the sample is small and deliberately chosen, we cannot generalize our findings

beyond this sample.

Data Collection

Interviews. Between March and June 2014, we conducted semi-structured interviews

with 142 teachers, administrators and other staff in the six schools. Interviews with

administrators lasted approximately 90 minutes and those with teachers approximately 45

minutes. At most schools, all members of the research team were present for interviews with the

principal, and with the directors of CMOs. In addition, all three researchers interviewed some

1 All names of schools, districts and individuals are pseudonyms. Some details have been changed to protect anonymity.

Ending Isolation

16

teachers at each school, which facilitated cross-site comparisons, improved inter-rater reliability

in coding data, and ensured that each research team member had informally observed elements

of every schools structures, practices, and culture.

We also purposively constructed our interview sample at each school. We recruited a

wide range of teachers, varying in demographics, teaching experience, preparation, and teaching

assignment. We solicited teachers participation by email, placing flyers in their mailboxes, and

following up on recommendations by the principal and other teachers we interviewed. In

addition, we interviewed key staff members (e.g. curriculum coaches, program and family

coordinators) when it became apparent that their work and views would inform our

understanding of teachers experiences. We provided participants with assurances of anonymity

and confidentiality.

In each school, we interviewed between 31% and 56% of the teachers, depending on the

schools size, the complexity of the organization and the practices used (For descriptive statistics

about the interviewees, see Appendix A). We used semi-structured protocols (Appendix B) to

guide our interviews and ensure that data would be comparable across sites and across

interviewers (Maxwell, 2012). All interviews were recorded and transcribed verbatim.

The interview protocols included questions that encouraged participants to discuss their

schools approach to teachers professional learning, including their work with colleagues. We

asked administrators whether their teachers were organized into teams and, if so, how those

teams functioned. In addition, questions about their overall view of the school provided

opportunities for participants to discuss the professional culture of their school. We probed

participants responses with follow-up questions.

Ending Isolation

17

Our reliance on interviews about a range of related topics limits the depth of information

we collected on any single topic, such as teams. However, this approach helped us to situate that

practice among other strategies for developing human capital and enabled us to consider

connections among them in our analysis. In our visits to the schools, we also observed a wide

range of day-to-day practices and looked for evidence about the schools organizational culture.

Document collection. Although interviews are the main source of data for this study, we

also gathered and analyzed many documents describing school policies and programs related to

recruiting, developing and retaining teachers, such as teacher handbooks, school policies,

meeting agendas, lesson planning templates, and data analysis forms used by teachers.

Data Analysis

After each interview we wrote detailed thematic summaries of the participants responses,

based on themes and findings from the literature on human capital as well as teachers work and

workplace. We next analyzed these the thematic summaries, identifying similarities and

differences across the sample as well as emerging themes in our study. We used these

preliminary analyses to supplement the etic codes drawn from the literature with emic codes that

emerged from the data (Miles & Huberman, 1994). Using this preliminary list of codes, we

reviewed a small sub-set of the transcripts, individually and together, in order to refine our

definitions and calibrate our use of the codes. After finalizing the list of codes and improving

inter-rater reliability, we thematically coded each transcribed interview using the software,

Dedoose (For a list of codes see Appendix C).

Having coded all interviews, we created data-analytic matrices (Miles & Huberman,

1994) to consider our research questions about teachers teams. We relied on Dedooses

function that allowed us to sort data by codes and by particular characteristics of participants.

Ending Isolation

18

We analyzed the data for each school separately, completing a number of data analytic matrices

and eventually cross-site analysis (Miles & Huberman, 1994.).

We used several strategies to address risks to validity. Throughout the process of

analysis and writing, we returned to the data to review our coding and check our emerging

findings, seeking rival explanations and disconfirming data (Miles & Huberman, 1994). Because

each of us collected some data in every school, we could compare and challenge one anothers

observations and emerging findings. We also asked colleagues who are familiar with this line of

research but not involved in this project, to review our drafts and offer alternative explanations of

our findings.

The State Policy Context

Although the schools we studied were located within one citys limits, each functioned in

a distinct policy context, which affected how that school functioned. Commonwealth Charter

schools, first authorized by Massachusetts in 1993, have the autonomy to choose their curricular

focus, allocate their budget, hire and fire teachers, and set their own schedule. In 2010, the state

also authorized local districts to approve Innovation Schools, which function as within-district

charter schools with many of the same autonomies. WCSD also permits schools to apply for

special status, providing them with the authority to, among other things, hire and transfer staff,

choose a curriculum, and develop their own schedule. As a result of these state and local

policies, in 2013 many schools serving Walker City students could exercise autonomy over

various aspects of their program and operations.

All students attending district or charter schools in Walker City were required to take the

MCAS annually, which was being aligned with the Common Core State Standards adopted by

the state in 2010. Based on students performance on the MCAS, the Massachusetts DESE

Ending Isolation

19

decided whether to re-authorize or close charter schools. DESE also rated all districts and

schools from Level 1 (highest rating) to Level 5 (lowest rating), based on their success in

meeting performance targets and narrowing achievement gaps among subgroups of students. If a

school rated at Level 3 or 4 failed to improve rapidly, despite assistance from the state, it could

be placed in turnaround status or closed. If placed in turnaround, the state required that a new

principal be appointed and that no more than 50% of the current teachers be rehired.

Alternatively, a failing school could be designated for restart by an independent operator. Both

types of schools could compete for three years of federal grants, administered by the state, to

support their improvement.

The Sample of Schools

At the time of our study, the sample of six high-poverty schools included three district

and three charter schools. However, this seemingly straightforward count masks a complicated

history that is important to understand. In 2009, four of the six schools in our sample were being

run by WCSD (Dickinson K-5, Fitzgerald K-5, Hurston K-8, and Kincaid 6-8) and two

(Rodriguez K-8 and Naylor K-5) were state-authorized charter schools. Then the state placed

two level 4 district schools (Hurston K-8 and Fitzgerald K-5) in turnaround status for failing to

improve. WCSD officials assigned each a new principal, who could select all certified staff,

potentially rehiring as many as half the current teachers. As turnaround schools, Fitzgerald and

Hurston received substantial government grants to fund their improvement plans as well as broad

managerial flexibility to choose staff, allocate resources, and schedule time. The following year,

the state intervened in a third failing district school (Kincaid 6-8), designating it for restart by a

charter management organization. When we collected data in 2014, all six schools (by then, three

district and three charter schools) had achieved the highest level (1) in the states accountability

Ending Isolation

20

system. Only one of the original district schools (Dickinson) had not been directly affected by

state intervention. These rapid changes in status, resources and flexibility for district schools

illustrate the impact of policies intended to equalize learning opportunities for students from

high-poverty, minority communities. Descriptive information about these schools appears in

Table 1.

The Charter Schools

Naylor Charter School and Rodriquez Charter School were well-established state charter

schools that opened in Walker City 10 years (Naylor) and 20 years (Rodriguez) before our study.

Initially, both were freestanding schools, although Naylor eventually became one of three

schools in the Naylor Charter Network. As Commonwealth Charter Schools, they were

completely free of all local district policies and were funded directly by the state.

The Kincaid Charter Network was selected to restart the failing Kincaid Middle School in

2011, three years prior to our study. Kincaid Charter School successfully recruited

approximately 80% of the students who had previously enrolled in Kincaid Middle School, a

higher proportion than typically returned to the school each year. Kincaid Charter officials had

invited the schools teachers to apply for positions in the new charter school; however, very few

applied and no one who did was hired. When Kincaid Charter opened, all teachers and staff were

new to the school. Because Kincaid functioned as an in-district charter school, the local teachers

union represented its teachers. Teachers pay aligned with the negotiated pay scale, but the

school was free from other provisions in the teachers contract, including those regulating the

academic schedule, the hours of teachers work, and evaluation practices. In addition, Kincaid

administrators controlled budget allocations, curriculum, and interim assessments. Within two

years, Kincaid Charter delivered significant gains in student test scores.

Ending Isolation

21

Table 1. Selected Characteristics of Six Sample Schools* School Name School Type Grades Estimated

Enrollment % Low-income

students

% African American

or Black Students

% Hispanic or Latino Students

% Other Non-white

students

% White Students

Dickinson Elementary

Traditional District PK-5 370 76 4 85 2 9

Fitzgerald Elementary

District - Former Turnaround

PK-5 390 85 70 25 3 2

Hurston K-8 District - Former Turnaround

PK-8 800 75 41 54 4 1

Kincaid Charter Middle

In-District Charter 6-8 475 88 50 30 10 10

Naylor Charter K-8 Charter K-8 500 82 70 24 5 1 Rodriguez Charter K1-8

Charter PK-8 420 72 55 20 7 18

*Percentages are approximated for confidentiality purposes.

Ending Isolation

22

The District Schools

Dickinson Elementary was a centuryold district school, serving a largely immigrant

student population, most living in the schools immediate neighborhood. For many years,

Dickinson was well-regarded within WCSD and recently had received commendation from the

state for its performance. The school experienced very low rates of teacher turnover; over half of

the teachers in 2013 had worked at Dickinson for more than 20 years. Dickinsons teachers were

represented by the local teachers union and the school was bound by the WCSD contract, as well

as state and district policies. Dickinson was a traditional district school, implementing district

policies and taking advantage of district resources, such as data coaches, curricular materials, and

interim assessments.

Hurston K-8 School and Fitzgerald Elementary School, also part of WCSD, had both

been placed in turnaround status as failing in 2010, four years prior to this study. At the

beginning of the turnaround process, the new Hurston principal replaced about 80 percent of the

schools teachers and Fitzgeralds principal replaced about 65 percent. Subsequently, both

schools demonstrated substantial growth on MCAS, allowing them to exit turnaround status at

level 1.

Although both schools continued to be part of WCSD after the state released them from

turnaround, they retained significant school-based control of their organization and management,

allowing them to maintain many of their reform initiatives. Both schools had autonomy in hiring,

budget allocations, scheduling teachers time, and choosing curriculum and assessments.

Teacher Teams as a Strategy for Improvement

Interviews conducted for our study provided ample evidence that teachers in all the

schools collaborated often about curriculum, pedagogy, and students. Administrators saw

Ending Isolation

23

collaboration as a means for improving teachers skills and unifying their efforts on behalf of all

students. Teachers said that collaboration helped them to manage the continuous, intense

demands of instruction and to align their efforts with those of their colleagues. Teachers in all

schools expressed respect for the knowledge, values, and hard work of fellow teachers. For

example, a teacher at Naylor Charter School said that the schools selective hiring process

ensured that she could rely on other teachers:

Everyones willing to help each other in the smallest way and to the biggest way. I feel like it works because those are the people that they hire. They look for someone who has a growth mindset. They look for someone who likes to work with people and can work successfully with people. If thats not someones waybecause everyones differentthey wouldnt survive here.

Similarly, a Hurston teachers praise and confidence in her peers echoed the comments of many

other teachers: [T]he amount of support we have as a staff, whether its from administration or

from each other, is amazing. We are a pretty cohesive group, by and large.

However, not all schools relied on formal instructional teams as part of their strategy for

turnaround or continuous improvement. Rodriguez Charter School encouraged teachers to work

together and assigned instructional coaches to support individuals who needed help. When

teachers preparation periods coincided, they could use that time to plan lessons together.

However, Rodriguez administrators did not expect teachers to meet regularly as teams and did

not arrange common planning time, which might have made interdependent work more likely.

Because the elementary and middle school units at Rodriguez were relatively small, individual

teachers often had no more than one colleague (and sometimes none) whose teaching

responsibilities matched their own. Rodriguez had one inclusion class at each elementary grade

level, which was co-taught by a regular and a special education teacher. Although these co-

teachers often said that they closely coordinated instruction, they did not collaborate closely with

Ending Isolation

24

teachers in other classrooms. During the schools weekly professional development meetings,

Rodriguez administrators often provided time for breakout groups of teachers to plan curriculum

or review students progress; yet teachers said they could not count on that time each week.

Elementary school Principal Rega, explained that time was scarce:

The big problem, honestly, is meeting time. These teachers are so stretched that to put another hour in to meet someone elsewe have a lot of meetings that people would like to have, like meetings with [the math and literacy coaches], and meetings with co-teachers, and meetings with grade-level teams, and meetings with parents. It goes on and on. . . . Its very hard to add another meeting.

Therefore, although Rodriguez teachers said that they collaborated with colleagues, they did not

report that a team oriented or supported their work.

In the other five schools, team meetings occurred regularly and teachers widely said that

their team helped them to teach better and contributed to their schools success, the two criteria

used by teachers to assess their teams in a prior study (Charner-Laird et al., 2015). Individual

teachers regularly described their reliance on their team. For example, when we asked a

Fitzgerald teacher leader with 6 years of experience whom she would go to for support, she

quickly responded, My team members. When asked further what kinds of support she might

seek, she answered, Everything. When we probed further about when this might occur, she

said, Every day, many times. We heard similar responses from supervisors and principals. A

Leader of Instruction at Kincaid characterized the team as the teachers first line of defense

saying, Its just a cohesive unit. . . . People are unified in their efforts here. You dont want to

see anybody fail. I definitely think that, more so than at my [prior] schools, teachers feel like

they can go to somebody and ask questions or admit if theyre struggling with something and get

support from their coworkers. I definitely think its a strength of this school. Fitzgeralds

principal, Sharon Forte, reflected on her schools rapid improvement after having been placed in

Ending Isolation

25

turnaround: I would say a lot of our success is because we really work at teams. The primary

unit is the grade level team. . . . Its really like you are married to your team. . . .

Consistent with our prior research about school-based instructional teams, we found that

teachers and principals judged teams to be effective when they met both individual and

organizational needs. Many teachers suggested that their teams enabled them to keep afloat

amidst the continuous, intense rush of academic and social activities of their school. Teams

increased their chances to succeed with their students, while continuing to contribute to the

schools improvement.

Teams Focused on Content and the Student Cohort

Although schools called their teams by different names, they essentially served two

functions: Content teams focused on developing curriculum, lessons, and pedagogy, as well as

monitoring student performance data (e.g., interim assessments, running records, exit tickets, or

unit tests) in order to gauge the effectiveness of their instruction. Cohort teams focused on the

students needs, behavior, and organizational culture. In some primary grades, the same group of

teachers met to address both matters of content and the cohort. However, in grades 5 through 8,

where teachers typically taught a single subject, separate teams focused on content and the

cohort. A Kincaid Charter School teacher described his schools team assignments: You

basically are always part of two teams. Youre part of a cultural [cohort] team, and youre part

of a department [content] team. Your department team teachers will never teach together, but

you will plan [your instruction] together. On your cohort team, you never teach the same

subjects, but you all teach the same kids.

Each of these five schools provided daily blocks of 50-60 minutes for teachers

preparation and development, which were scheduled simultaneously as common planning time

Ending Isolation

26

for all teachers in particular grades or content areas. During at least one of these blocks weekly,

sub-groups of teachers met as teams to address academic content, the cohort, or both. Teachers

could use their remaining blocks of preparation time to work independently or meet with

colleagues.

Content Teams Planned Instruction and Monitored Students Learning

All five schools had content teams, composed of teachers who taught the same subject(s)

either within a grade or across grades. This included primary grade-level teachers who taught

multiple subjects in self-contained classes and teachers of upper-elementary or middle school

students, who taught a single subject in several classes or grades. Sometimes content teams also

included special education or English as a Second Language teachers, especially when those

individuals taught regularly in an inclusion class.

The schools in our study relied less on prepared curricula than their counterparts in many

traditional schools, although the district schools (Dickinson, Fitzgerald, and Hurston) used some

of the curriculum provided by the WCSD. Teachers in these five schools were not expected or

encouraged to plan their curriculum and lessons on their own. Instead, collaborative planning

occurred in the context of content teams.

For example, Hurston elementary teachers met with their grade-level team to plan the

sequence of topics and competencies they would all teach, which then guided their decisions

about curriculum units and daily lesson plans. Second-grade teachers, who taught all subjects in

self-contained classes, reported having spent their grade-level content meetings during the prior

10 months planning reading units that aligned with the Common Core State Standards. The team

leader called this a very daunting task. We never had a common curriculum for reading.

Previously, the team had done similar work in math:

Ending Isolation

27

Every week, the team would talk about the next topic of math and figure it out together, trouble-shoot problems we foresaw and think about ways to maybe [provide] more hands-on experiences [for students]. That felt healthy, just sitting down as a team and talking about it week to week, and figuring it out. Dickinsons teachers in third, fourth, and fifth grades regrouped their classes for English

Language Arts (ELA) and math so that each teacher specialized in a single subject and taught all

students in the grade. During content meetings with the principal, the teachers reviewed their

progress, planned content for future units, and monitored their students success in mastering the

standards addressed in the units. These teachers also met frequently during common planning

time and lunch to review their day-to-day plans, reflect on their lessons and discuss the progress

of students they shared.

Experienced and new teachers routinely reported that their grade-level and content teams

reduced the uncertainty about what and how to teach by enabling them to meet the continuous

demands of planning, teaching, assessing, and revising a curriculum that stretched throughout the

school year. A Kincaid history teacher with nearly a decade of experience summed up the

benefit of his content teams process: [It has] helped turn the job of curriculum design into a

much more manageable beast.

Integrated lesson planning. Whereas content teams at Dickinson K-5 and Hurston K-8

shared the task of developing curriculum topics and units, those at Fitzgerald K-5, Kincaid

Charter, and Naylor Charter took collaboration further by sharing responsibility for writing

lessons. This involved a complicated process that required everyones participation and included

tight weekly deadlines. Typically, curriculum planning started with state standards, which teams

studied in order to identify the knowledge and skills that students would need to learn. Based on

that, they designed instructional units. Individual teachers then took responsibility for writing

lesson plans for a particular subject or day, which all team members then used in teaching the

Ending Isolation

28

unit. In some cases, lessons were very detailed, including scripted introductions, explanations

and questions that the teachers could use to promote deeper thinking.

Before teaching any lesson, team members at Kincaid reviewed, critiqued, and often

revised the proposed plan. One team leader explained,

The first 45 minutes of the planning block, you review their lessons, add comments, and then we go through lesson by lesson and kind of talk through what revisions need to be made, where it looks like its strong, where it needs to be improved.

However, content meeting time also served broader instructional purposes for many teams. One

7th grade ELA teacher at Kincaid described her content teams meeting earlier that day:

We planned out our students culminating essay for the unit that theyre going to work on. We talked about what we want it to look like. We looked at exit tickets from this past week to look at the data of what it was showing us about their progress. We talked about what book we want to teach in the next unit, and then we looked at each others lessons for next week.

As grade-level teams met at Fitzgerald K-5, teachers of self-contained classes each took

responsibility for planning lessons for one subject, which were then taught by all grade-level

teachers. A second-grade teacher explained, Theres three of us [at the grade level] this year. I

do reading, another teacher does the math planning, and another teacher does the writing

planning. Those are our core subjects. Principal Forte decided which teacher would take

responsibility for each core subject, based on her judgments about their different levels of

expertise. She also required teachers to use a uniform planning template. A third-grade math

teacher showed us his lesson plans, explaining that each included the objective, state standard,

materials, key vocabulary, mastery of facts, activating background knowledge, questions to ask,

learning and strategy activity, opportunities to practice, assessment, and closure. By Thursday

of each week, Fitzgeralds teachers were expected to upload their draft lessons for the following

Ending Isolation

29

week so that their colleagues and the school administrators could review them. Team members

provided feedback during team time, while administrators sent theirs in writing.

At Naylor Charter, teachers on content teams also planned approximately five lessons

each week, which they and their colleagues then used to teach their classes. One teacher with 10

years of teaching experience said, Right now Im planning math. . . One of my co-teachers is

planning reading. . . and then another teachers planning all of the science and writing.

Whereas Fitzgeralds teachers were assigned to write lessons for the same subject throughout the

year, Naylors teachers changed assignments, themselves. As one explained, We shift it. . . . I

started planning with reading. This is my second time planning math. I planned writing once. I

planned science for a while. So we just shift, usually with the units.

Most teachers at these three schools spoke very positively about integrated lesson

planning. Arguably, preparing 20 lessons each weekas teachers of traditional, self-contained

elementary classes often mustlimits the level of planning that any teacher can do. Fitzgeralds

Principal Forte said, The struggle is how do you keep people from burning out? Teams, she

said, served as a major part of her schools response. However, teachers we interviewed at

Fitzgerald offered less consistently positive assessments of the shared lesson planning process, in

part because the principal assigned teachers responsibility for particular subjects and required

them to use a detailed, standardized template. One who spoke positively recalled how he and his

colleagues initially responded to the requirements: We complained a bunch. However, he said

that the administrators sold it by convincing teachers that they could tweak, adjust, [and]

differentiate the initial lessons in subsequent years. They said, Youll thank us. That

actually turned out to be true, thankfully. He also thought that the lessons were sufficiently

flexible so that teachers could use their professional judgment: The principal made it very clear

Ending Isolation

30

that teachers have their own styles. Even though you get a plan from me that is written in a very

specific way, you need to meet that objective in the best way suited for you. However, another

teacher thought that this process of lesson planning was not taken seriously and claimed that few

teachers actually used the lessons. More often, however, teachers spoke positively about shared

lesson planning but expressed misgivings about the principals expectations for consistency,

which sometimes eliminated instructional options that some teachers might have preferred.

Although interdependent lesson planning was meant, in part, to protect teachers from

burnout, teachers were not all convinced that it reduced their work overall. For example, a math

teacher with seven years experience at Naylor said,

Co-planning is hard. Its one of the reasons that people like working in schools like this and then also get burnt out working in schools like this. . . . So much of the planning [still] comes down to the teacher, making the materials, designing stuff themselves. It just is a reallyits a lot of work.

Notably, several of these schools maintained on-line banks of units and lessons, sometimes

accompanied by videos of the lesson being taught. Teachers who drew upon these resources said

that they made their work easier and their instruction better.

Several teachers saw the advantage of integrated lesson planning, but worried about their

loss of autonomy. One explained that interdependence has both advantages and disadvantages:

What doesnt work well is kind of the same thingthat its shared. So you lose a little bit of control or a lot bit of control sometimes over what is going in front of kids. Last year, where we had a team member who was not very strong at curriculum design and didnt seem to like it very muchhis lessons were not very good. It was frustrating in the sense that, through this process, unless the person . . . really wants to do well, it can have a kind of negative effect.

Noting that the teacher had left the school, he said, This year its working much better.

Another teacher raised a similar concern. [T]he biggest con to working collaboratively

is that sometimes the way you want to teach something is not the way the other person wants to

Ending Isolation

31

teach. . . . Its a problem to figure out common ground every single week. Its rough and tough

and exhausting. In these schools such differences either were eventually resolved or dissatisfied

teachers leftor were asked to leavethe school.

Surprisingly few of the many teachers we interviewed said that they lacked confidence in

their colleagues ability to plan curriculum or write lessons. One called her fellow teachers rock

solid, an assessment that many others seemed to share. Teachers who expressed doubt usually

had a particular colleague in mind, leading them to conclude that teams worked well when all

teachers were qualified and committed. Recognizing this challenge, one teacher said,

Thankfully, at least [with] the teams that Im on, people are all on board.

Still, even principals acknowledged that integrated lesson planning had some limitations.

Principal Forte observed that although its benefits were many, Its not perfect. Because when

you dont write your own plan, you dont know it as well. Kincaids Principal Kain, was

concerned about the challenges of implementing Common Core Standards: Especially in

English, the curriculums so sophisticated, its really hard to negotiate it with three people. Im

not fully convinced that works.

Monitoring students academic progress. All schools in the study dedicated some

team time to analyzing data about students academic achievement, which helped them gauge

their instructional success. Naylor and Fitzgerald scheduled separate meetings for data analysis,

while Dickinson, Hurston, and Kincaid incorporated data analysis into content meetings. For

example, at Dickinson, Principal Davila centered her weekly team meetings on data analysis.

Together, she and the teachers from two grade levels (k-1, 2-3, and 4-5) reviewed results from

state tests, interim assessments, or assignments created by teachers. A district-sponsored data

coach attended meetings twice each month to work with Davila and her teachers as they

Ending Isolation

32

reviewed student achievement data. At Naylor, this work occurred in smaller subject-based

teams. A math teacher described how his 7th-grade content team redesigned their unit on

probability after students performed poorly on a new interim assessment that was aligned with

the Common Core. The depth of what they were being asked about probability increased pretty

substantially. He said teachers on his team werent happy with the way the kids were

performing on those types of questions and MCAS too. . . . There was one MCAS question that

just destroyed our kids. It was something that we just had overlooked. He and his colleagues

realized that they needed to change both the content and the ways we delivered it.

Cohort Teams Monitored Students Needs, Behavior, and Organizational Culture Whereas content team meetings explicitly focused on curriculum and instruction, cohort

team meetings were dedicated to ensuring that students could and would do their part as learners

and citizens of the school. Teams did this by systematically discussing the needs of individual

students within their grade-level cohort, reviewing the groups behavior and compliance with

rules, and strengthening aspects of the organizational culture. For example, one of Kincaids

resident teachers who partnered with a middle-school cohort described a typical Friday team

meeting, where teachers discussed what happened during the week, what students were doing

well, what they were not doing well, whether any individual students had problems.

Reviewing individual students. All schools designated some team time for monitoring

students needs and behavior, which was said to increase teachers shared sense of responsibility

for all students. Teachers believed that focusing on individual students during cohort meetings

ensured that no students personal needs or performance would be overlooked. At Dickinson

and Fitzgerald, this was largely done at grade-level meetings, which included both content and

cohort discussions. However, each of these schools also convened a standing student support

Ending Isolation

33

team that met weekly to review the success and needs of individual students throughout the

school. These two schools viewed students academic performance to be inextricably tied to

experiences with poverty, their home life, and their socio-emotional health. Other schools

focused less on the conditions of students lives than on their behavior within the walls of the

school. Teachers at Hurston K-8, Kincaid Charter, and Naylor Charter met in separate grade-

level cohort teams where they reviewed students, both individually and as a group. By

focusing on the academic and personal well-being of individual students, teachers could identify

those who were experiencing difficulty in several classes and then intervene to get them back on

track. For example, Hurstons middle school cohort teams met with relevant student support staff

(the dean of discipline and counselor for the grade) for two consecutive blocks each week. One

teacher described the process for reviewing individual students:

Its as easy as, Hey, can we put Felix on the agenda for Friday? It might start as an email early in the week. It could start like that or a casual conversation between teachers. Then wed decide what has to happen from there. Is it a conversation with a teacher who has a relationship with him? Is it a phone call home? Sometimes we invite parents to come up during these meetings. We might set aside time. If the counselors there, they might recommend a course of action. . . .

These interventions were generally effective, but sometimes required longer-term focus on

particular students over time. If we do revisit, its usually like, Hey, this isnt working. . . .

The students that weve presented, typically, whatever issue it is gets ironed out.

Monitoring students behavior. Each of the schools had detailed rules about students

dress (for example, uniforms at Naylor; uniforms at Kincaid with shirts tucked in; uniforms and

no sandals at Hurston); how they should behave in the corridors (for example, silence at Naylor

and Kincaid, quiet filing at Fitzgerald and Hurston); and how they should behave in classes

(respectfully, attentively in all schools). The teachers sought to be explicit about their

expectations and consistent in their responses and rules were designed to promote an orderly

Ending Isolation

34

environment, conducive to learning. During cohort team meetings, teachers took stock of their

students compliance with rules and expectations. Given the schools various philosophies, some

focused more than others on compliance. For example, as a no excuses school, Kincaids

teachers regularly reviewed individual students adherence to the schools standards for behavior

and responded quickly to violations. Teams in several other schools focused more generally on

whether the rules were being upheld by the group.

Strengthening the cohorts culture. Teams also created new activities, incentives, and

rewards to motivate students in their cohort. They sought to establish and nurture a distinctive

organizational culture, intended to keep students enthusiastically and intently focused on learning.

For example, Hurston middle school had adopted a set of behavioral norms known as PRIDE

(Perseverance, Respect, Integrity, Daring, and Excellence). Teachers recognized individual

students when they acted in a way that was consistent with a PRIDE norm by giving them a

small certificate. One teacher explained, It was just in a moment, You did something good;

here you go. You have to earn 19 more of these to [qualify for] the ice cream party at the end of

the month. Explaining that this had a more positive tilt than reactive punishment, the teacher

also noted that the top 80 percent of the class gets to go.

At Kincaid, where each grade level was divided into three cohorts, a teacher explained

that within her cohort We all run our classrooms the same way. The same expectations for the

kids. We have the same consequences. We have the same incentives. We have the same cheers,

the same chants. They also created new activities, incentives, and rewards to motivate students

in the cohort. The team leader of one explained that certain practices were consistent

throughout the school, while others were customized for the cohort. To illustrate, she said that

students were expected to call out answers or raise their hands in similar ways across the school,

Ending Isolation

35

but the prizes that were giving for highest homework completion might be customized by

cohort.

The Factors that Contributed to Teams Success

Despite considerable school-to-school variation in how teams worked, a similar set of

factors emerged that participants suggested contributed to teams success: a worthy purpose and

continuous focus on learning; sufficient, regular time to meet; engaged support by

administrators; facilitation by trained teacher leaders; and integration of teams with other efforts

to support and develop teachers. We next explain how each worked and why it was important to

the teams success.

Having a Worthy Purpose and Continuous Focus on Learning

Teachers resent being compelled to commit scarce time to tasks that lack meaningful

purpose and practical payoff (Charner-Laird et al., 2015). Based on her research in non-profit

organizations and for-profit companies, Edmondson (2012) concludes that managers (principals,

in this case) support effective teams by framing an aspirational purpose for their ongoing

interaction (p. 100). In contrast to the unproductive team meetings often described in the

literature, participants in this study suggested that their work together was fueled by such a

purpose. Each of these schools had an explicit mission to eliminate the racial achievement gap

and reduce educational disparities between wealthy and low-income students. Their mission

informed key decisions, beginning with recruitment and hiring and continuing through

observations of teaching, and reviews of students progress. The schools mission gave purpose

and practical meaning to teachers work on teams as they developed curriculum and lessons,

assessed the success of their instruction, and monitored students behavior, needs, and progress.

A Naylor administrator explained that the way they achieved their mission was to give our

Ending Isolation

36

students the education they deserve. These schools did not simply have a mission; they were, as

Principal Forte told her teachers, on a mission.

Hargreaves (1994) distinguishes between schools where teachers engage in genuine

collaboration and those where they participate in what he calls contrived collegiality. He

explains that in schools with true collaboration, team meetings are voluntary and do not occur

solely at a fixed time or in a predetermined place. They are oriented toward development, and

may lead to unpredictable outcomes. By contrast, in schools where collegiality is contrived,

meetings are compulsory, occur only at scheduled time and predetermined place, focus on

implementation, and are meant to have predictable outcomes (p. 208). In our study, teachers

experience with teams did not align fully with Hargreaves list of criteria for either genuine

collaboration or contrived collegiality. Team meetings certainly were not voluntary, although

very few teachers suggested that they felt coerced to attend. It was simply how they did their

work together. Many explained how team time contributed to their own learning and to their

students learning through defined, coordinated processes. In that way, these teams might be

viewed as being both development-[oriented] and implementation-oriented. Although

meetings were regular, they were not bounded and fixed in time and space. Instead, the

teachers work together spilled over into times before school, during lunch, and after school,

when they conferred together. Also, although administrators expected teachers to use their time

together productively and some tracked what the teams accomplished, they did not insist on

uniform predictable outcomes and often encouraged creative solutions to challenging problems.

Overall, these teachers would not have characterized their work on teams as contrived

collegiality, although they also would not have said that it was guided entirely by the teachers

interests and preferences.

Ending Isolation

37

Sufficient, Regular Time for Team Meetings

These schools all gave team time priority over many other activities. Anticipating regular

meetings allowed teachers to know that it would be worth their time to plan an agenda and arrive

ready to follow through with it. Other studies have found that, when team time is short,

intermittent, or unpredictable, meetings often become occasions for check-ins with colleagues

about routine mattersplayground monitoring or field trip attendanceor opportunities for

individuals to socialize or prepare for their next class (Troen & Boles, 2012; Neil & Johnston,

2005; Supovitz, 2002). The schools in this study widely treated team time as inviolable.

Although each of these schools dedicated blocks of time for teams to meet, their

arrangements varied widely across the sample, depending on the time that was available during

the teachers work day and whether a teachers contract limited the principals say in how that

time might be used. The following examples from Dickinson K-5 (with the least dedicated time

and principals discretion) and Naylor Charter School (with the most time and principals

discretion) illustrate the range of approaches in these five schools.

Team time at Dickinson K-5 and Naylor Charter. As a traditional district school,

Dickinsons teachers were guaranteed by contract to have five blocks of time each week for

preparation and planning. Principal Davila scheduled those blocks so that teachers had common

planning time across two grade levels (K-1, 2-3, and 4-5), which allowed them to meet in smaller,

horizontal or vertical sub-groups if they chose. It was no simple task to arrange for these classes

to be covered simultaneously by specialists teaching art, music, or physical education. The

WCSD contract authorized the principal to decide how teachers used one block each week.

Although this was far less than other principals controlled, Davila made the most of it. Every

Tuesday, she convened and facilitated three sequential team meetings for teachers from each pair

Ending Isolation

38

of grades (K-1, 2-3, and 4-5). As she explained, they plan together and collaborate and talk

about individual students, interactions that typically extended beyond a single block each week:

They have four common planning times, and if I walk around during their common planning

times, theyre usually sitting in each others classrooms, talking about planning. Teachers

confirmed that they spent considerable time during their preparation periods, at lunch, and after

school working with colleagues.

At the other end of the spectrum was Naylor Charter School, where the teachers

workday was 8 hours long, and team time was much more extensive and closely regulated by

administrators. Teachers at Naylor were not unionized and therefore no contract specified how

they could use their non-teaching time. Every day, teachers met at least once with three or four

teachers from their grade-level content team to develop curriculum and lesson plans. The same

content teams met for another block each week to review data about their students performance.

Then each Friday during the schools professional development time, grade-level teams met

again for one to two hours. Teachers also met weekly as cohort teams to review the progress of

students they shared and to discuss discipline and culture within their cohort. Samantha Nelson,

director of the Naylor charter network, explained that the network had big expectations for

collaboration. . . One of our four big organizational values is We grow best together. The

schools generous allocation of team time reflected that belief and reinforced that priority.

Finding the time. Administrators at other schools did their best to protect and dedicate

regular blocks of time for team meetings. As heads of charter schools, administrators at Naylor

and Kincaid could create their own schedule for professional development and team meetings. At

Hurston and Fitzgerald, teachers met as teams during two or more planning periods, rather than

only one as the districts contract allowed. Most teachers welcomed the fact that state and

Ending Isolation

39

district regulations granted their school the additional autonomy that made that possible. In fact,

Fitzgeralds Principal Forte reported that teachers voted to continue using two preparation

periods for team meetings after the school emerged from turnaround status.

Providing predictable time for collaboration proved to be essential, both to make

collaboration possible and to convince teachers that school leaders believed their work together

was important. A Hurston teacher was impressed with the fact that the time itself is even

carved out. . . . Her colleague concurred:

We now have schedules that are like a dream. Its arranged so that our content and grade-level meetings are two consecutive hours a week. That uninterrupted time is so precious that we can actually get a whole lot of work done. . . . This year, the new piece is that the grade-level teams all meet at the same time. That way, if we need to check in about something or two teams might need to come together on something, we can do that.

Therefore, although Principal Davila successfully implemented teacher teams at Dickinson,

despite constraints on the time and schedule, the experiences of the other four schools suggest

that principals with more time to work with and a greater say in how it would be used found it

easier to ensure teams effectiveness. Across these five schools, however, when schools secured

enough time for teams to participate in serious, sustained work, teachers willingly invested in the

process.

Engaged Support by Administrators

Administrators at all these schools were invested in their teams success and monitored

their progress, either up close or from a distance. At a minimum, they emphasized the

importance of the teams work and followed the plans and decisions that teams made. Some

principals attended meetings often, although Davila was the only one who chaired her schools

team meetings.

Ending Isolation

40

Hurstons administrators each took responsibility for following teachers work on four

content teams. Principal Hinds assignment included English Language Arts content teams at

three grades and the school wide Arts team.

Those are my four teams. So I go to almost all of their meetings. . . . All of us, the administrative team, are on the Google Docs for all the teams and the Listservs for all the teams so that we can follow electronically whats happening, even if were not there.

A middle school teacher leader in math said:

[Principal Hinds] doesnt micromanage, but he plays a role in some of the small decision-making we have in our different teams. Hell pop up and attend different team meetings or hell read the notes and give feedback. But its not Okay, you guys. You have to do this. This team, you have to do this. This team, you have to do that.

A second-grade teacher confirmed the principals account:

We have a [teacher leader] as team facilitator. Its one of us. He leads the meetings. Theres also an administrator assigned to attend every meeting. . . . Hes there more just to keep it on track and suggest and answer questions when they come along that none of us on the teaching level really know that answer.

However, several teachers said that when ELA test scores in grades 3-5 failed to improve over

time, Hinds stepped in and, as one said, laid down the law, requiring the team to focus on skills

featured by the Common Core, such as close reading.

Kincaids administrators, like Hurstons, sometimes took a more active role if a team

encountered difficulties or students test scores fell. A Leader of Instruction in math said, Our

seventh grade team, I go to almost every planning block just because theyve struggled more,

just in terms of getting results in student achievement. Im there just to provide extra support to

them. She contrasted that with her involvement with the sixth- and the eighth-grade teams,

where I pop in every once in a while, because the teams are just really strong and the leaders are

running the show really well, so I tend to not prioritize being at those.

Facilitation by Trained Teacher Leaders

Ending Isolation

41

One of the most notable features of teams in the three schools where the state had

intervened (Hurston, Fitzgerald, and Kincaid Charter) was that teacher leaders formally

facilitated them. When they were assigned to turnaround status, Hurston and Fitzgerald both

partnered with Teach Plus, a national non-profit organization that selected, trained and supported

teacher leaders for their schools. Hurston committed a full-time administrator to supervise

teacher leaders work. This individual met weekly with each teacher leader to review the prior

team meeting and help to plan for the next.

With very few exceptions, Hurstons teachers praised their team leaders, although not

surprisingly, some who held the role found it challenging. For example, a teacher leader who

had been at the school since it entered turnaround, said Its gotten easier over time. . . . We have

a group of high-performing, outstanding, all-star teachers from everywhere. . . . Being tasked

with leading that team was daunting. . . He recalled reflecting, Well, I dont know everything

in ELA. How can I possibly do that? With help from his supervisor, he came to realize that

it wasnt about having all the answers, but rather about being able to steer folks towards the

answers. . . . The support here [from my administrative supervisor]. . . has made all the

difference in the world.

Teacher leaders at Kincaid offered similar views. A content team leader, who met

weekly for supervision with his Leader of Instruction, said that he was still trying to understand

his role, which straddled a line between manager and colleague: We work as a team, but I am

a point person, and they do come to me in regards to suggestions about the sequence, and

brainstorming content, and so forth. He said that the opportunity for constant

discourse . . .[was] super enriching, and fulfilling for me as a teacherknowing that Im making

an impact, not only in my classroom, but Im affecting the entire grade. A sixth-grade teacher

Ending Isolation

42

at Kincaid who had taught only three y