Strengthening Partnerships

How Communication and Collaboration Contribute to School Improvement

For most of the past decade, I have studied union-management efforts to improve public education. In this time, I have witnessed extraordinary examples of teachers, union leaders, and administrators who are working together to improve teaching and learning. These examples provide a counterstory to the policies that seek to reform education through the use of markets—specifically, charter schools and vouchers—or through the use of high-stakes testing as a way to evaluate teachers and improve instruction.

Reforms based on market forces and testing take school improvement in the wrong direction, yet these ideas have dominated the policy debate over improving public education. In this debate, teachers and their unions have often been characterized as the problem, not part of the solution. What is missing in all the discussion is a systems perspective on improving public schools that examines the way schools are organized, the way decisions are made, and the way teaching and learning are improved.

Before starting my career in academia, I spent nearly 10 years as a consultant to unions and management that were trying to improve their quality and productivity performance in the face of increasing global competition. While unions and management have conflicting interests around certain elements in the employment relationship—for example, the division of profits—they also certainly have common interests in making the organization as strong as possible with high-quality products and services. So working together around those common interests made sense. I was also interested in the "productivity of democracy," the idea that more democratic organizations, ones that value employees' voice and offer them more decision-making opportunities, can be more productive. When employees are allowed to contribute meaningfully to solving problems and making decisions, better solutions are found, and those solutions are implemented more effectively because people are more committed to solutions they have a hand in developing.

This approach toward improving productivity and quality included developing systems of extensive employee participation. It involved creating more-collaborative team-based organizations that could plan, identify, and solve problems; make decisions; and implement solutions. I've been fortunate to work with unions and management in a variety of industries undergoing these extensive changes due to globalization, including auto, steel, electronics, telecommunication, aerospace, pharmaceutical, and paper-making—and now education.1 My first experience extending this participative approach to public schools dates back to 1988 in Altoona, Pennsylvania, where I saw the potential of collaborative arrangements.2

Through those years consulting, I learned that collaboration between labor and management around common interests such as quality can be a tremendous competitive advantage. I also saw how unions as democratic institutions can mobilize a workforce in ways that management cannot achieve on its own. Union leaders who are elected by members have a level of greater trust that management on its own cannot generate. Unions mobilizing members and providing independent leadership can add tremendous value to organizations by finding new solutions to problems and implementing those solutions effectively.3

I also witnessed that, while organizations could develop team-based work systems and joint union-management partnerships to improve problem solving, decision making, and performance, sometimes those changes were undermined by the forces of international markets. So, I went back to graduate school to learn more about globalization. My doctoral work focused on General Motors' Saturn Corporation, an excellent example of a union-management partnership that achieved levels of quality never before seen at GM.4 I became convinced that this type of industrial democracy could lead to a competitive advantage as well as a much more satisfying and fulfilling work experience for all employees, because they had more voice in identifying and solving problems, and shared in managerial decision making and firm governance.

After many years researching these partnership arrangements in a variety of industries, I was invited to join a study tour of unions in the United Kingdom that Sandra Feldman, then the president of the American Federation of Teachers, and officials from the Albert Shanker Institute were taking. Our visit there focused on another form of industrial democracy—the learning representative system. In this case, union representatives in the U.K. were taking on the responsibility for helping members improve their skills and knowledge through additional training and education. This was my first introduction to the AFT, and I became aware of a number of excellent examples of long-term union-management partnerships that were transforming education in local school districts. These partnerships fostered collaborative approaches to curriculum development, scheduling, budgeting, strategic planning, hiring, K–12 subject articulation, interdisciplinary integration, mentoring, professional development, and evaluation. I became interested in whether a set of underlying patterns existed that was common to all these partnerships that had been sustained for more than a decade. This led to the research in which I am still engaged today.

Where Partnerships Are Strong

In the first stage of this research, my doctoral student John McCarthy and I studied seven cases of collaborative partnerships between teachers' unions and administrators who had been working together in innovative ways to improve teaching quality and student performance for more than a decade.5 The school districts included:

  • ABC Unified School District in Cerritos, California;
  • Toledo Public Schools in Ohio;
  • Hillsborough County Public Schools in Florida;
  • Plattsburgh City School District in New York;
  • Norfolk Public Schools in Virginia;
  • Independent School District 15 in St. Francis, Minnesota; and
  • Charlotte County Public Schools in Florida.

These districts are located across the country and are a mix of urban and rural, large and small. A number of AFT leaders helped us identify these districts and locals based on their strong reputations for having institutionalized a long-term collaborative partnership between the district administration and the local union focused on school improvement, teaching quality, and student achievement. AFT staff often accompanied us on our visits. So these cases were not selected randomly, and I do not claim they are a representative sample of all districts nationally. Rather, they are examples of what is possible when unions and administrators develop collaborative partnerships to manage and improve a school district. As such, they are worthy of study to see what can be learned from their stories.

Our research team visited all of these districts and interviewed union presidents, school board members, superintendents, central office administrators, principals, union representatives, executive board members, teachers, support staff, and members of the community. In addition, we studied their collective bargaining contracts, memorandums of understanding, student performance data, and relevant internal reports.

We then analyzed these cases to identify the themes and patterns that were common to all these districts. The themes fell into four broad categories:

  1. Motivation for initiating collaboration
  2. Strategic priorities for improvement
  3. Supportive system infrastructures
  4. Sustaining characteristics

1. Motivation for Initiating Collaboration

In almost all of these cases, a crisis or some pivotal event helped motivate a change in union-management relations. In most cases, a strike, or a vote to strike, was the critical event that prompted the districts to seek a new direction in their union-management relationships. In doing so, they recognized that the adversarial relationships that led to the strike or vote were counterproductive and not in the best interests of teachers, administrators, or students. Union and administrative leaders in each district made the choice to change their relationship, which was the first step in establishing a collaborative partnership in school improvement.

2. Strategic Priorities for Improvement

In their union-management partnerships, all districts emphasized joint work on strategic priorities, including teaching quality and student performance. To that end, they all engaged in substantive problem solving and innovation for improvement. For example, districts developed union-led professional development, new systems for teacher evaluation, teaching academies, peer-to-peer assistance, and mentoring programs. The result for most of these districts was very low levels of voluntary teacher turnover.

These districts also created multiple opportunities for teachers and administrators to work jointly on analyzing student performance data in order to target areas for improvement. School-level partnerships facilitated collaboration on developing data-based improvement plans. Teachers also formed teams at the grade or department level to use student performance data to guide improvement efforts. Partnership districts reported high levels of student achievement, including in schools with high percentages of students receiving free or reduced-price lunches.

Through these partnerships, districts engaged in substantive innovation and experimentation around areas critical to student achievement and teaching quality. Some examples of these innovations include the joint establishment of:

  • Reading programs in schools with high percentages of students receiving free or reduced-price lunches;
  • Peer assistance and review programs;
  • Systems for teacher evaluation that measure student growth;
  • Teacher academies focused on professional development;
  • Curriculum development; and
  • Sophisticated systems for analyzing student achievement data to better focus student interventions.

The partnerships are clearly vehicles for system improvement, not ends in themselves.

3. Supportive System Infrastructures
Culture of Collaboration

Most of these districts have created an organizational culture that values and supports collaboration. In this "culture of collaboration," school district administrators promote trust and value the leadership that the union brings to the entire district. Both union and management leaders speak of a culture of inclusion and involvement, as well as the importance of respect for teachers as professionals and for their union. Collaborative planning, problem solving, and decision making are embedded in the way the district is managed.

Shared Governance and Management

Additionally, these districts have established a model of shared governance, in which formal joint planning and decision-making forums allow the union and administration to work together and align the strategic priorities of the district. They have also developed an infrastructure that gives the union significant input in planning and decision making around issues such as curriculum, professional development, textbook selection, school calendar, and schedules. The act of managing is viewed as a set of tasks that leaders (both union and administration) must engage in for the benefit of teachers and students. As a result, "management" is not viewed simply as a separate class of employees.

Collaborative Structures at All Levels

Collaborative structures are found at all levels in these districts. Such structures allow district and union officials to promote and facilitate collaborative decision making at the school level through forums such as building teams, school improvement committees, school steering committees, leadership teams, or school advisory councils that meet regularly. These bodies are vehicles for planning and decision making around issues such as goal setting, budgeting, policymaking, discipline, and safety.

Union as a Network

In these partnership districts, collaboration extends beyond top district administrators and union officials into the school buildings themselves. Data teams, grade-level teams, and department teams are led by union members who participate in substantive decision making around curriculum, instructional practice, and K–12 articulation. Further, most of these districts have developed peer-to-peer mentoring programs to support professional development courses that involve teachers as teacher-leaders, master teachers or mentors, and professional development trainers. When we look at the numbers of union members involved in district- or school-level committees or teams, along with the numbers of teachers involved as mentors, teacher-leaders, master teachers or professional development trainers, in many cases they add up to more than 20 percent of the union membership, which is a high percentage of members involved in union activities compared with most locals. This results in a dense union network, meaning that professional relationships between union members and administrators, and among union members themselves, are very strong and allow for open lines of communication for improving teaching and learning. The "denser" or stronger the network, the better the district can solve problems and implement new initiatives or programs rapidly and effectively with a great deal of support. This union-based implementation network is something managers report they could not create on their own. It also institutionalizes the partnership in the district by embedding collaboration in the way the district is managed.

Social Capital and Joint Learning

These partnership districts have invested heavily in creating opportunities that allow for joint learning and for building the professional capacity of administrators and teachers. In such districts, opportunities exist for union and management leaders to learn together through shared educational and training experiences. These experiences, in turn, allow for knowledge acquisition (the development of human capital) while also strengthening working relationships (the development of social capital).

In some cases, the development of both human and social capital has meant inviting hundreds of union leaders and administrators to attend planning retreats within school districts; AFT-sponsored events such as the TEACH conference, Center for School Improvement training, Union Leadership Institute programs, offerings from the Professional Development Program for Educators; university-based educational programs; and corporate leadership programs. Since these experiences are shared between the union and management, leaders from both hear the same message and get the same information at the same time. Through this education, they can experience each other as colleagues with mutual interests who can work together to improve teaching and learning.

4. Sustaining Characteristics

Long-term leadership plays an important role in districts with strong labor-management partnerships. In most of these districts, the local unions have been led by longtime union presidents—some who have led for as long as several decades. Many of these districts have also enjoyed long-term leadership from their top administrators. Continuity of leadership provides stability for these district-union partnerships, and also allows for effective working relationships to be formed directly between the union president and the superintendent. Most of these superintendents came from the districts themselves, with some having served as teachers and union members before joining the administration. Such steady internal labor markets support the culture of collaboration by allowing trust to be built between leaders who have known each other and worked together for years.

These districts have also recognized the importance of engaging the community. They have involved community members or parent groups in school-based governance structures or in district-level planning processes.

In many districts, after deciding to engage in greater collaboration with management, local unions have also become increasingly involved in school board elections by recruiting or supporting specific candidates. In some cases, they have helped defeat candidates who did not support a partnership approach to school governance. These locals have realized that since the school boards hire the superintendent, electing board members who support collaboration will increase the chances of finding willing partners in administration.

These local unions and districts have also received support and resources for their collaborative partnerships from the national union. AFT President Randi Weingarten has made collaboration in school reform and improvement a high priority:6

More and more, ... our leaders are building strong relationships with school administrators, doing the hard work of collaborative school improvement—and producing better results for children. ... The one thing [partnership districts]  all have in common is a culture of collaboration—a universal recognition among business leaders, public officials, community leaders, parents, and teachers that they can accomplish great things for students if they work together.

This support has translated into several types of AFT-sponsored technical assistance, including the conferences and training I mentioned earlier, as well as resources from the AFT Innovation Fund, which supports initiatives for union-led innovation and improvement.

In some cases, unions and districts have negotiated contract language (see "Community Building in Cincinnati") or memorandums of understanding that support their collaborative efforts so as to institutionalize the partnership. For instance, these contracts may call for collaboration in district-level decision making by requiring union representation on key committees. In other cases, enabling language in contracts has expanded opportunities for union involvement in decision making through school  board policy that promotes inclusion in professional development programs, textbook selection, hiring, peer assistance, mentoring, and teacher academies.

What We Have Learned

These examples of collaborative school reform represent an alternate path in the debate over education policy. This path views schools as systems; it allows for a focus on improving and restructuring public schools by the people working in the schools themselves to improve planning, decision making, problem solving, and the ways teachers interact. The districts described here demonstrate how teachers and their unions have been vital to improving public education systems in collaboration with administrations.

To highlight these examples, the AFT, in collaboration with scholars from Rutgers University's School of Management and Labor Relations, Cornell University's School of Industrial and Labor Relations, and the Sloan School of Management at the Massachusetts Institute of Technology, and with funding from the Bill & Melinda Gates Foundation, organized the first National Conference on Collaborative School Reform, which was held in October 2010 (see the box on page 6). This event was a precursor to the conferences later sponsored by the U.S. Department of Education. Thirty-five district teams of union leaders and administrators from across the country came to Washington, D.C., to learn from these examples and the relevant research, and to discuss how models of collaboration, which contribute to school improvement, might be pursued in their own districts. Conference participants heard presentations from teams representing the ABC Unified School District in Southern California; Plattsburgh, New York; Toledo, Ohio; St. Francis, Minnesota; Norfolk, Virginia; and Hillsborough County, Florida. They explained how deep, sustained partnerships have resulted in true shared decision making at the district and school levels, new support networks for innovation and instruction, and data-informed decision making in schools.7

These union leaders and managers discussed how partnership efforts were created and sustained over the past two decades. In 2011, the research from the conference was published in a policy report by the Center for American Progress, Reforming Public School Systems through Sustained Union-Management Collaboration, by Saul Rubinstein and John McCarthy.8

Next Phase of Study: Collaborative Partnerships and Student Achievement

At a time of increased focus on student performance data, we felt that an important but underexplored area of study within union-management partnerships was the relationship between collaboration and student achievement. So, we are currently examining the patterns of collaboration that occur within schools among teachers and administrators, and looking to see if and how they affect student performance.9 As a first step, we collected data from the ABC Unified School District and the ABC Federation of Teachers in California. This district has 30 schools and more than 900 educators. Forty-six percent of its students qualify for free or reduced-price lunches.

To understand the impact of partnerships and school-level collaboration on student performance, we used partnership attitude and climate surveys, data from the California Academic Performance Index (API), and social network analysis. The API includes standardized test results in math, English, social studies, and science as well as graduation and dropout rates. Social network analysis explores whom teachers and administrators communicate with on a regular basis, how they communicate, and what topics they discuss. From this we can model the patterns of communication within and between individuals and schools.

We measured partnership quality at the school level using questions from a districtwide partnership survey in 2011 that dealt with union-management communications, collaboration among staff, and openness to input from all educators. Communication network data were drawn from a social network survey administered in 2011. In this network survey, educators were asked to indicate the other teachers and administrators they communicated with, and, specifically, if they communicated to:

  • Discuss student performance data;
  • Discuss curriculum development, cross-subject integration, and articulation;
  • Share, advise, and learn about instructional practices; and
  • Give or receive formal and informal mentoring.

The density values for these networks are calculated as the proportion of existing communication links in a school, to the total possible in the school.


Using our 2011 survey data on the quality of school partnerships, and analyzing those data against 2011 and 2012 student performance data, we were able to examine the relationship between the strength of the partnership and both the level of API performance in 2012 and the difference in student performance between 2011 and 2012. First, we found that partnership quality bears a positive and statistically significant association with overall API performance in 2012. A 1-point increase in partnership quality in 2011, based on a survey with a scale of 1 to 4, corresponded with more than a 25-point gain in API scores in the 2011–2012 school year after controlling for poverty. On average, this represents an increase of 3 percent in API for each 1-point gain in partnership quality. Further, we found that partnership quality in 2011 bears a positive and statistically significant association with performance improvement from the 2010–2011 school year to the 2011–2012 school year. For example, a 1-point gain in partnership quality in 2011 corresponded to a roughly 15-point gain in API scores over the following year.

When we analyzed the relationship of school-level partnerships (based on the partnership survey) with school-level collaboration (based on the network survey), we found that those schools with the strongest partnerships also had the highest levels (density) of teacher-to-teacher communication, meaning that more teachers discussed student performance data, curriculum, articulation, instructional practice, and mentoring with one another in stronger-partnership schools than in weaker-partnership schools.

Figure 1 illustrates the difference between the "density" of collaboration—the strength and number of professional networks among teachers—in stronger- and weaker-partnership schools. For purposes of illustration, this chart combines the densities of the four communication topics—student performance data, curriculum and articulation, instructional practice, and mentoring—and compares the density (percentage of teachers communicating regularly with each other) in the strongest- and weakest-partnership schools. As you can see from the chart, teachers in the stronger-partnership schools have almost twice the communication density as the weaker-partnership schools. As with the relationship between partnership quality and API performance, this association was statistically significant.

[[{"type":"media","view_mode":"wysiwyg","fid":"848","attributes":{"alt":"Figure 1: Frequency of Teacher-to-Teacher Communications","title":"Figure 1: Frequency of Teacher-to-Teacher Communications","height":"308","width":"414","style":"height: 308px; width: 414px;","class":"media-image media-element file-wysiwyg"},"link_text":null}]]

In addition, we found a very interesting difference in the structure of union-management relations in schools with strong partnerships when compared with weaker-partnership schools. As shown in Figure 2, building representatives in strong-partnership schools tended to have more frequent and less formal communication with their principals than did building representatives in schools with weaker partnerships. Again, this difference is statistically significant. This finding illustrates the changes that occur through union-management partnerships at the school level. Institutional union-management partnerships take place between the union as an institution and the school district, but individual partnering also takes place between the superintendent and union president, and between the principal and building representative. What we see in the ABC school district is a strong association between the frequency and informality of communication between the building representative and the principal, and the quality of the partnership from the perspective of the teachers.

[[{"type":"media","view_mode":"wysiwyg","fid":"849","attributes":{"alt":"Figure 2: Frequency of Communication between Building Representatives and Principals","title":"Figure 2: Frequency of Communication between Building Representatives and Principals","height":"639","width":"365","style":"height: 639px; width: 365px;","class":"media-image media-element file-wysiwyg"},"link_text":null}]]

In summary, our research leads us to conclude the following:

  • The quality of formal partnerships between teachers' unions, administrators, and teachers at the school level had an important and significant positive impact on student performance as well as performance improvement, even after controlling for poverty.
  • High-quality teacher-administrator partnerships predicted "denser" school-level collaboration and communication (stronger and more numerous networks between teachers and administrators) around: (a) student performance data; (b) curriculum development, cross-subject integration, or grade-to-grade integration; (c) sharing, advising, or learning about instructional practices; and (d) giving or receiving mentoring.
  • Strong-partnership schools have structurally different patterns of union-management collaboration. The strength of partnerships predicted different communication patterns between union building representatives and principals, with the communication in high-partnership schools becoming more frequent and less formal.

*  *  *

Our research explores the significant and important impact that school-level union-management institutional partnerships can have on teacher collaboration and student performance. We find that partnerships offer an often overlooked, yet highly relevant, story in policy debates over public education. Our research suggests that unions can and do play a leading role in school improvement by partnering with administrators to improve teaching and learning. Such partnerships add tremendous value to school districts seeking to improve and sustain high levels of student achievement.

Union-management partnerships are a sound alternative to narrow, market-based testing and accountability strategies. Tests can reveal shortcomings in knowledge, but not how to deal with those knowledge gaps. In contrast, union-management partnerships are designed to create solutions for improving teaching and learning. Indeed, partnerships provide the basis for a network facilitating teacher-to-teacher and teacher-to-administrator communication and collaboration, which are crucial to school improvement.

Innovations in collaboration will not be replicated or sustained, or become institutionalized, without widespread support from state and federal policy. In light of our findings, we suggest that policymakers provide schools and districts with incentives for collaborating on evaluation, mentoring, professional development, and peer assistance and review programs, and also allow for waivers or mandate relief from unproven testing, accountability, and market reforms for districts implementing collaborative reform efforts.

These partnerships take work and are difficult to sustain. But once a culture and system of collaboration is institutionalized, great results do emerge. We hope these findings and experiences will help other districts and local unions that want to pursue a strategy of collaborative school reform. And we hope this research will encourage policymakers to design incentives for greater collaboration among teachers' unions, administrations, and boards of education.

Saul A. Rubinstein is a professor in the School of Management and Labor Relations at Rutgers University and the director of its Program on Collaborative School Reform. His research has focused on labor-management efforts to transform industrial relations, work systems, and performance in a wide variety of industries. This article is adapted from two reports by Rubinstein and John McCarthy, written for the Center for American Progress: Reforming Public School Systems through Sustained Union-Management Collaboration (2011), and Teacher Unions and Management Partnerships: How Working Together Improves Student Learning (forthcoming).


1. Saul A. Rubinstein, "The Local Union Revisited: New Voices from the Front Lines," Industrial Relations: A Journal of Economy and Society 40, no. 3 (2001): 405–435.

2. James A. Kimple, Saul A. Rubinstein, Dennis Murray, and Edward Blair, "Applying Quality Principles in Public Education," Annual Quality Congress (Milwaukee, WI: American Society for Quality Control, 1991).

3. Saul A. Rubinstein, "Unions as Value-Adding Networks: Possibilities for the Future of U.S. Unionism," Journal of Labor Research 22, no. 3 (2001): 581–598.

4. Saul A. Rubinstein, "The Impact of Co-Management on Quality Performance: The Case of the Saturn Corporation," Industrial and Labor Relations Review 53, no. 2 (2000): 197–218; and Saul A. Rubinstein and Thomas A. Kochan, Learning from Saturn: Possibilities for Corporate Governance and Employee Relations (Ithaca, NY: Cornell University Press, 2001).

5. Saul A. Rubinstein and John E. McCarthy, Reforming Public School Systems through Sustained Union-Management Collaboration (Washington, DC: Center for American Progress, 2011); and Saul A. Rubinstein and John E. McCarthy, "Public School Reform through Union-Management Collaboration," Advances in Industrial and Labor Relations 20 (2012): 1–50.

6. Randi Weingarten, "Conflict Makes Good Stories, Collaboration Makes Good Schools," Huffington Post, January 5, 2011.

7. American Federation of Teachers, "Conference on Collaboration Details 'Reform from Within,' " press release, October 12, 2010,

8. Rubinstein and McCarthy, Reforming Public School Systems; and Rubinstein and McCarthy, "Public School Reform."

9. Saul A. Rubinstein and John E. McCarthy, Teacher Unions and Management Partnerships: How Working Together Improves Student Learning (Washington, DC: Center for American Progress, forthcoming).

Reprinted from American Educator, Winter 2013–2014


Related Articles

Supporting Labor-Management Collaboration

Cultivating Collaboration
The Science behind Thriving Labor-Management Relationships
By Greg Anrig

Banding Together
Union City's Teachers and Administrators Work Together to Improve Student Achievement
By David L. Kirp

Strengthening Partnerships
How Communication and Collaboration Contribute to School Improvement
By Saul A. Rubinstein

Moving Meriden
In Connecticut, a Road Map for Union-District Relations
By Jennifer Dubin

The Bargaining Table and Beyond
How the AFT Came to Support Labor-Management Collaboration
By Phil Kugler

Download the Article (1014.85 KB)
American Educator, Winter 2013-2014