From small schools to powerful tools for school improvement: A conversation with Mark Dunetz about the evolution of New Visions for Public Schools

In 1989, New Visions for Public Schools was founded on the belief that public/private partnerships and small school designs could help improve the drop out rate in New York City schools. Since then, the organization has gone through several iterations, expanding from a focus on incubating small schools, supporting a network of small schools, and most recently developing data-based tools that can support the work of educators across and within schools. To better understand this evolution and the issues that New Visions is working on today, we spoke with Mark Dunetz, the Vice President for School Support at New Vision and previously a principal at a New Visions high school.


New Visions 1.0: An Incubator For The Development Of Small Schools

The early evolution of New Visions for Public Schools reflected the growing interest and commitment to the development of new small schools that swept the United States in the 1990s. In fact, while New Visions began with a grant from the Carnegie Corporation to create an after school program to engage students in community service, by 1993 they had received a $25 million grant from the Annenberg Foundation to create 14 new small schools in New York City. In 1996, they officially adopted the name New Visions for Public Schools in order to reflect their focus on supporting the development of wide variety of teacher- and student-centered small secondary schools. With grants from the Gates Foundation and other funders, New Visions has gone on to open 140 schools in New York City. Interestingly, although divisions are often made between those who support opening new public schools and opening new charter schools, New Visions has done both. As the number of charter schools in New York City and elsewhere has grown throughout the 2000’s New Visions has opened up seven charter schools. As one of the few organizations bridging the “charter divide,” several of New Visions’ charter schools have been opened in cooperation with the teachers’ union and Michael Mulgrew, current President of the local union, sits on the New Visions Board. Counting both the public schools and charter schools that New Visions has opened, today, 1 in 5 students in New York City schools attend a school either opened or currently supported by New Visions.


New Visions 2.0: Supporting the Development of a Network of Schools  

New Visions’ expansion from starting new schools to supporting and sustaining a network of schools took advantage of significant changes in local educational policies made after New York City Mayor Michael Bloomberg took office in 2002. Bloomberg, along with School Chancellor Joel Klein, sought to dismantle what they saw as an inefficient and overly-centralized school bureaucracy, by releasing schools from direct oversight by area superintendents. In contrast, they sought to grant local schools and principals with autonomy while holding them accountable for improvements in student performance. A central element of their approach was the creation of a competitive environment in which School Support Organizations tried to attract schools that would pay for their services. Rather than remaining centered on the creation of new small secondary schools, New Visions embraced this opportunity to become a School Support Organization and began to work with both new and existing schools on day-to-day operations. As a result, New Visions expanded to work with both large and small schools, as well as transfer schools (schools designed to re-engage students who have dropped out or who have fallen behind in credits) and grade 6-12 schools. Dunetz described this as a “laboratory for doing deep work on the day-to-day of everything happening in schools.”


New Visions 3.0: Creating Tools to Innovate Across the System

With another change in school administration in New York City, New Visions has expanded its focus again. In 2015, following the election of Mayor DeBlasio, Carmen Fariña was appointed the new Education Chancellor for New York City schools. She ushered in changes to the organizational structure of the system that included re-establishing the role of local superintendent and eliminating the competitive marketplace for School Support Organizations. However, because New Visions along with a few other organizations was regarded as an effective network-provider, it was allowed to retain its network of schools. As a result, New Visions still works with a select group of schools; however, the support is no longer provided in a marketplace environment. Schools can opt to be members of the New Visions PSO Network for a three-year period. As Dunetz described it, this allows New Visions to play out a set of strategies and focus more on innovation and less on competition. While there was always an explicit expectation that New Visions would work on a set of strategies that would in theory have value for the larger system, in some ways they are also back in a place where they can serve as an innovation lab and incubator for the development of tools and practices that address some of the key problems that their schools and others face.

Today, as Dunetz explained, many schools face challenges implementing their visions. From his perspective, efforts often fail at the point of implementation because people can’t organize quickly and efficiently enough to carry out anything substantially different. As he put it, “I’ve seen that innovations don’t last over time because people can’t keep up with it and can’t work out the details of something non-traditional.” Furthermore, he argues that success may be more likely to happen in places with select or non-representative student populations – where students are screened and the organization and the normal pressures of traditional schooling may be reduced. According to Dunetz, “The places that have tried to implement innovation with a typical non-selective population and with large numbers of students, they tank academically and wind up regressing towards common practice.”

In order to combat some of these challenges, New Visions’ current work focuses on finding ways to use technological tools to target areas in need of support in schools—tools developed from his own experiences as a principal. For example, New York City has a complicated set of graduation requirements that makes it very difficult for schools to keep track which students are making progress at an appropriate rate. Traditionally, every principal and school has had to figure out how to solve this problem on their own. However, New Visions worked with their schools to develop a scheduling tool that makes it possible to see whether students are enrolled in the appropriate classes and gaining the credits that they need. As Dunetz described the issues and their current approach:

I experienced, along with my colleagues, that you could do a whole lot with simple technology. Tools have become one really big piece of our strategy. We no longer see it as a necessary evil, where you have to go out and find a vendor that is the least bad. We see the control of the development of those tools as a very powerful mechanism for changing behavior.


Where we shifted a year and a half ago was to a much more explicit modeling of what it looks like to use tools at the administrative level. Now we’ve got to go to other levels. We do what we call  “strategic data check ins.” These are scripted, protocol-driven conversations that look at key planning tasks organized around the tool, or multiple tools. We do it largely through Google hangout, so we can do a large number of schools over a period of time. We go through and come up with plans that are recorded alongside the data, and then we pull back and look across so we can be a second set of eyes. That’s become a core part of our strategy and it’s become tremendously successful at shifting practice at scale very quickly around very high stakes things, like what constitutes a meaningful graduation plan and what are the smartest strategies for sitting and preparing students for Regents exams. We can organize systematically. That’s a huge step forward for us.


Today, New Visions is focusing on the notion that in order to be effective at regular planning you need a common reference point that is updated and available to everyone involved. The organization is working on designing a framework that takes information out of the heads of the many individuals who work with students, and puts it on paper. Dunetz described this information as more specific than generalized information, but not prescriptive for solving problems. “It’s the guts of the system,” he explained. “It’s what needs to happen in order to be able to sustain innovation, and to be transparent. See all moving pieces and what is and isn’t being implemented with fidelity. It’s different than what people are used to. People are used to a highly prescriptive checklist. Our hope is to get schools to a level of functioning on a whole set of things that can be solved in a short period of time.” With these developments, New Visions now has over twenty staff members working on data analytics and designing systems and structures that can be used by their schools and others.

All in all, New Visions has expanded from starting small schools, to incubating small schools, to leading a network of schools. Now, it serves as one example of a new kind of educational organization that goes beyond school design and school support to develop tools and practices that meet the day-to-day needs of teachers and principals in schools of all kinds.

Deirdre Faughey

A conversation with Eric Schwarz about the evolution of Citizen Schools

Reflecting on incremental and disruptive change in school and out: A conversation with Eric Schwarz about the evolution of Citizen Schools

Eric Schwarz

Eric Schwarz

I had a recent conversation with Eric Schwarz about the evolution of Citizen Schools. Eric is the founder of Citizen Schools and author of The Opportunity Equation: How Citizen Teachers Are Combating the Achievement Gap in America’s Schools. Citizen Schools began in 1995 in Boston as an afterschool program designed to provide opportunities for students in low-income communities to participate in apprenticeships with mentors from a wide range of professions. Since that time, it has grown into a national organization that partners with public middle schools to integrate apprenticeships and other rich learning opportunities into a longer school day. While Citizen Schools works only in the United States, the evolution of the Citizen Schools model illustrates broader questions of educational change and innovation that are relevant around the world. In this post, you can listen to the interview, get a summary my conversation with Eric, and see slides(below) I developed for my class on school change that highlight issues of “incremental” and “disruptive” change and innovation. See weeks 2 and 3 of the online syllabus for related references and resources, including readings by David Tyack and Larry Cuban, and Clayton Christensen.  

Citizen Schools 1.0: “Apprenticeships are core”

When Schwarz first started working on the ideas that led to Citizen Schools, his key concerns included the limited access that many students in lower income communities have to the kind of rich and extended learning opportunities commonplace in many upper income communities. In order to address this problem, Eric and his colleagues created an afterschool program that engaged students from several middle schools in low-income communities in activities like producing a newspaper in which they worked with volunteer mentors – “citizen teachers” – from related professions. The program took place in the schools, for a few hours a week, with a little time devoted to tutoring and help with homework, but with the main focus on the apprenticeships.

Schwarz explained that in the early years, the work was challenging, but the feedback from many parents and students was powerful. At the same time, some of the same parents who saw these benefits also raised concerns that there was no corresponding improvement in most students’ grades or academic performance. In some cases, parents wanted to take their children out of the program because they felt that they needed more help with their school work. In addition, principals who were providing Citizen Schools with space also wanted to see more academic benefits. As Schwarz put it in The Opportunity Equation, with increasing pressure on schools and principals from new state policies demanding improved performance, some principals had “less tolerance for our rookie mistakes and seemed in some cases to lose their appetite for the enrichment-based learning we were offering.” (p. 63)

Citizen Schools 2.0: Supporting academic development

In response to the feedback they were getting, Schwarz and his colleagues decided to refine the Citizen Schools model. They wanted to remain focused on apprenticeships, but also chose to make support for academic development a more explicit goal and aspect of the design.  To do so, they started offering the program on an almost daily basis, substantially increasing the amount of programming they could offer. They continued to offer the apprenticeships but also significantly increased the time they spent working with students on homework and tutoring. In order to staff those additional hours, however, they also had to change their staffing model and, in addition to the Citizen teachers, they brought in AmeriCorp volunteers to work in the program on a regular basis, and they also hired program directors to work on full-time rather than half-time. With these changes, students continued to report powerful experiences, but the grades and test scores of many of the students improved as well. As Schwarz explained, “The schools changed us in a good way–they made us better academically.”

Citizen Schools 3.0: The extended learning time edition

Citizen Schools continued to expand its afterschool model in Boston and then to a few other cities, but then in 2006, Citizen Schools had the opportunity to partner with The Edwards School as part of a pilot program in Massachusetts to support schools in developing Extended Learning Time (ELT) models. The approach to ELT developed at the Edwards School built on the Citizen School’s approach, but it also had to respond to local circumstances at the Edwards that influenced their design and that made it different from many other ELT models. In particular, rather than simply adding the Citizen Schools afterschool program onto the end of the school day, they integrated the apprenticeship approach and the added help with academics into the regular school day. As a consequence, the school was able to offer a 2 hour “elective/apprenticeship block” four days a week, add a regular 60 min “math league”, and provide time for a half-day of professional development for teachers every Friday. To make that possible, Citizen Schools provided the staff for many of the apprenticeships/electives, but Citizen Schools staff also took on responsibility for teaching some academic subjects, particularly, math. While progress was slow at first, eventually there were clear signs of significant improvements. In addition to offering increased instructional and enrichment programs, there improvements in test scores including an 80% reduction in achievement gap between students’ performance in ELA and science on state tests. Furthermore, while only 17 families in 2005 made the Edwards School their first choice on the application asking which middle school they would like the children to attend, in 2008 over 450 families applied to the school (see The Opportunity Equation, p. 92).

Citizen Schools 4.0: Spreading the model and reshaping thinking about “after” school

With the development of the ELT model and some success both at the Edwards School and in other schools where they tried the ELT approach, questions of how to scale the model came to the forefront. Correspondingly, one key part of the work since that time has been to build an organization capable of spreading the model to other schools. However, as Schwarz explained, the current model is very intensive, requiring significant staffing, time and resources. As he put it, it might be possible to develop the model in a hundred schools, but probably not a thousand. In response, Citizen Schools has also begun exploring other options to support further spread. In particular, they have started to work more broadly in collaboration with government agencies and policymakers to create programs and resources that can support ELT models and afterschool programs in general. Further, they are partnering with private and public organizations and networks to advocate for the development of more ELT models and to develop a broader vision for how learning time can be used both in schools and after school.

The development of Citizen Schools to this point illustrates the ways in which changes and innovations in education evolve between and among organizations and schools. In this case, the Citizen Schools model adapted in response to the challenges and opportunities encountered when working with and in schools during after-school and in-school time. While the integration of the Citizen Schools approach into the regular school day may not have disrupted many of the conventional structures and patterns in schools, learning opportunities for students have expanded. Further, one can argue that Citizen Schools has, at least to some extent, “disrupted” conventional views of after school programs and now provides a different model for afterschool programs as well as an existence proof that that model can be effective. At the same time, the efforts to disrupt the conventional institutions, structures, and beliefs that reinforce inequalities in access to educational opportunities continue. While Eric has stepped down as CEO of Citizen Schools, which is now led by Steven Rothstein, he plans to turn his attention to new work, perhaps in higher education.

Slides: hatch citizen schools
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

The evolution of the Millennium Villages Project: Building infrastructure & local expertise

Dr. Rahika Iyengar

Dr. Rahika Iyengar

Over the past year, IEN has begun a series of interviews and events focusing on educational innovation around the world. While attention often centers on how “innovations” and other interventions are supposed to change teaching and learning, some of our interviews will highlight how the work and ideas of leading thinkers and organizations have evolved in response to the realities they have encountered. (For example, see an interview with Eric Schwarz about the changes that Citizen Schools made in order to successfully grow their model for afterschool and extended learning time and an interview with Brahm Fleisch on school improvement efforts in South Africa.) This week, we draw on a conversation with Radhika Iyengar, Director of the Education Sector at the Center on Sustainable Development at the Earth Institute at Columbia University. That conversation concentrated on how the educational initiatives of the Millennium Villages Project (MVP) in Africa have evolved along with efforts to address numerous challenges in agriculture, in health, and in other sectors.

The Millennium Villages Project was launched in 2005 to demonstrate how to achieve the Millennium Development Goals in a number of economically marginalized areas in rural Africa through “integrated, community-led development.” The project was designed to take into account the complexity of problems like poverty and to address them through inter-sector efforts to work on agriculture, health, infrastructure, education and other issues. As Iyengar put it “everything had to come together.” The work on education in particular illustrates the way work in one sector – like improving teaching and learning in classrooms – depends on and can draw from work in other sectors. In order to improve access to quality education on a large scale, the project had to deal with the facts that infant and child mortality was often high, that some children did not have adequate transportation to get to school, and those who did make it to school might be malnourished or in poor health.

Given the problems with nutrition and health, the MVP began with a focus on addressing agricultural issues in several different parts of Africa. The initial goals included improving local food supplies and making sure that farmers could be successful. Work was soon launched to tackle challenges in other sectors as well. In each case, however, “root causes” for complex problems needed to be unraveled in order to develop integrated, systemic solutions. In health, for example, the project quickly focused on key issues including the high numbers of children dying from malaria and large numbers of mothers dying during childbirth. While providing vaccines could help to address infant mortality issues, the health workers soon found critical problems with the supply-chain created by the lack of infrastructure in many regions. Thus, in order to retain their effectiveness, the vaccines needed to be kept cool when transported and stored. That created a need for refrigerators which in turn, required electricity or other power sources that were unavailable in many of the areas where the vaccines were needed most. As a consequence, pursuing the health issues, quickly merged with efforts to develop infrastructure in the villages, including establishing electricity and building health centers.

With each development, however, new challenges were uncovered. Once the health centers were built, considerable work needed to be done to make sure that the local population would take advantage of the new facilities and services. “Just creating health centers on their own was not enough,” Iyengar explained. “Mothers would not go to the health centers on their own. They were too busy. They had many other things to take care of. They had their agricultural work, their children, husbands, the household.”   Convincing mothers who were used to delivering at home to make the trek to deliver their children in an unfamiliar setting was particularly challenging.

In response to these new challenges, the MVP developed a group of community health workers, individuals from the local community who would work with the local population to help them take advantage of the new clinics, hospitals and other services. These individuals could communicate in the local languages and dialects and understood the local beliefs and norms. They visited houses, checking for malaria, testing for HIV, talking to pregnant mothers, educating mothers, checking blood pressure and assessing whether anyone needed to go the hospital. By working closely with the community health workers and a team of local officials and community members, staffer from the MVP were also able to gain key information, feedback, and perspectives that helped to identify issues and generate solutions that fit the local context.

Although these challenges in other sectors had to be addressed, they also created some unanticipated resources, lessons, and opportunities that could then be leveraged to work on other issues of education. Thus, the work on education in 2005 began with a focus on increasing access that in many ways paralleled the work in health. The challenges for increasing access to preschool and kindergarten were particularly pronounced as younger children often could not walk the long distances it might take to get to the nearest schools. Furthermore, many of the existing schools were dilapidated and in disrepair. Consequently, the work in education began by developing the physical infrastructure and building more schools and classrooms. The work proceeded with the MVP providing materials and assisting local teams who built new schools and made existing schools more functional. But as Iyengar explained, simply building more schools and classrooms was not enough to ensure that students would get to schools and classroom at the right time.

Communication was a key factor as many parents simply didn’t know what age students should come to school. Some students did not start coming to school until they were 7 or 8 years old; but sometimes students as young as 3 or 4 might tag along with their older siblings. Similarly, many families did not have information about when school started and ended or the timing of school vacations.

Building on the success of the Community Healthcare Workers in bridging the gap between households and the health centers, the MVP created groups of volunteers they called the Community Education Workers. These local team members went house-to-house discussing the school schedules and policies and identifying children ready for school (including helping families to identify when children were born and how old they were).

Through these kinds of efforts, more and more children started attending school on a regular basis. For instance, on average, of the total children enrolled in grade 3 in all schools in all sites, more than 85% now attend schools on a regular basis. As in many other countries, simply getting students through the door and into classrooms was not sufficient, however. There were still teacher shortages, overcrowded classrooms, and teachers struggling to teach large numbers of students from different grades and working on different subjects at the same time. Given the difficult conditions, even though more students were attending schools, many were dropping out. “They were either getting bored,” Iyengar elaborated, “or parents were seeing that after three or four years the students still weren’t able to read and write.” As Iyengar stated, “there are many, many policy issues, structural issues, instructional issues, implementation issues that we are still grappling with in terms of quality of education.”

One strand of the work to improve the quality of education has focused on developing the capacity to share and use information and data more productively. For the most part, educators and policymakers in the MVP communities have had little or no information on what’s actually happening at scale, across many different schools and classrooms in real time. To address this challenge, the work in education borrowed again from the efforts in health and the community healthcare workers. The healthcare efforts have taken advantage of the development of the technological infrastructure to use a variety of new technologies to carry out testing and to share information between households, local villages, hospitals and the regional centers. That information is being used to make individual diagnoses and treatment decisions and also enable healthcare workers to manage demand and supply for medicines, doctors, and hospital beds. These efforts benefited particularly from the developments in mobile technology and from equipping community healthcare workers with smartphones. In education, they have also begun using smartphones as well. In this case, community education workers use the phones to gather a wide range of data including data on the school environment as well as on student learning. For school environment the data may be as simple as “During the visit, was there electricity? Was there running water?” but the education data includes information on progress in reading and numeracy as well. As Iyengar points out, however, simply transmitting the data to some distant location for eventual analysis and reporting is often of little use to the teachers, teams, and local officials. Thus, the developments in the use of the mobile technologies and assessment at the school level, now go hand in hand with the development of monthly meetings where the local data can be examined, discussed and used to guide decision-making. For example, this data was used to identify and address problems with teacher absences and with low levels of reading in Mali and Ghana respectively. From Iyengar’s perspective, there are often relatively simple solutions that local teams can identify, but solving the problems depends on both on the development and use of the new technologies and the development of social communication networks that link communities with one another and with the regional and national education officials.

Part of the challenge that MVP faces is common to all those working in schools: efforts to increase efficiency and improve performance at one level of the education system – classroom or school – are still constrained by conventional policies and practices at other levels. Thus, in many cases, the educational issues that the MVP works on are also complicated by the histories and conventions of the national education systems. For example, the regional or national languages like French or English often serve as the language of instruction. But many students and teachers do not speak the language of instruction and books, textbooks, and other instructional materials are often not available in the local dialects. As a consequence, educational improvement efforts like those of the MVP face the dual challenges of trying solve problems that are created by the conventional structures of schooling at the same time that they work to increase access and improve quality within those conventional systems. (For more on these issues, see reflections on and discussions of innovation/improvement in Finland, Mexico and Colombia, and Singapore.)