From hunger aid to school reform

From hunger aid to school reform: An interview with Jerry Sternin

Positive Deviance Approach seeks solutions that already exist

By Dennis Sparks

Journal of Staff Development, Winter 2004 (Vol. 25, No. 1)

JSD: I'd like to clarify some of the underlying beliefs that support the value of the Positive Deviance Approach. But before we do that, what is a positive deviant and what value does he or she provide to an organization?

Jerry Sternin: Positive deviants are people whose behavior and practices produce solutions to problems that others in the group who have access to exactly the same resources have not been able to solve. We want to identify these people because they provide demonstrable evidence that solutions to the problem already exist within the community.

Deviants everywhere

JSD: You are saying that within virtually every group--which I assume means virtually every school--there are individuals whose behavior enables them to get better-than-average results and that these individuals have discovered pathways to success for the rest of the group. That's contrary to a particularly potent and usually unquestioned assumption in the field of education--and perhaps in other fields as well--that the solution to problems must come from the outside because those on the inside either don't know enough to improve things or don't have the will to do the hard work of change. That view is particularly strong regarding schools that have a history of low performance.

Sternin: That is a prevailing view in other fields as well. For example, in our work on malnutrition in Vietnam, experts told us that malnutrition was caused by food distribution patterns, illiteracy, poor sanitation, the role of women in society, and so on. But we found parents in the villages who hadn't read the textbooks and didn't know they couldn't have a well-nourished child until those problems were solved.

Outsiders

JSD: You are quoted in Fast Company as saying, "The traditional model for social and organization change doesn't work. It never has. You can't bring permanent solutions in from outside."

Sternin: My experience in over 12 years of working with this particular approach and more than 30 years of experience in the development field is that improvement may occur when an external agent brings new resources and ideas to a community. But as soon as that external agent leaves, the problem returns because the recipients were essentially passive. This is why best-practices approaches usually fail.

Discoveries change behavior

JSD: You believe that it's critical that group members discover new ways of doing things rather than having a solution imposed on them. And yet school leaders often say that it doesn't make sense to continually reinvent the wheel.

Sternin: People learn best when they discover things for themselves. Knowledge is usually insufficient to change behavior. It is our own discoveries that change behavior. A basic belief of the Positive Deviance Approach is that when someone from the outside provides the solution, those to whom it is directed may not believe it and do not have an investment in it. Once that person leaves, it is difficult to sustain the changes.

It's important to understand, though, that the Positive Deviance Approach is always used within the context of a very specific problem. We wouldn't try to improve educational performance because that is too broad a subject. Instead, we would define within it particular problems and do a Positive Deviance Inquiry for each of them. For instance, parents' participation in their children's education would be a separate inquiry from improving reading in the primary grades. This makes it easier to identify the positive deviants, who may be different individuals in different situations.

Don't tell me what to do

JSD: I can imagine at least some school leaders saying, "But our teachers don't want to discover. They just want to be told what to do. Then they resist taking the advice."

Sternin: It's natural for people to resist when someone tells them what to do. That's part of human nature. It's like the human immune system's rejection of anything it senses as foreign. It's the same thing at the psychological and emotional levels when an external solution is imposed on us. But when the solution comes from within the system, the immune response isn't activated.

Change behavior first

JSD: You emphasize the importance of transferring behavior or practice, not knowledge. A great deal of professional development in schools is information transmittal, or what I call the passing of lists--10 things a principal can do to be an instructional leader, five ways teachers can better manage their classrooms, eight attributes of an effective mentoring program, and so on.

Sternin: In positive deviance work we say that it is easier to act your way into a new way of thinking than to think your way into a new of acting. So our focus is on changing behavior. We are not interested in people listing good nutritional practices; we want them to use them.

In the development world the conventional wisdom is that knowledge changes attitudes and attitude changes practice. Positive deviance reverses that. We start with changing practice. As people see that those changes make a difference, their attitude changes and they internalize the knowledge. We can spend our lives learning about something, but that doesn't necessarily change our behavior.

A Positive Deviance Inquiry focuses on behaviors, not individuals. We look at all the uncommon but demonstrably successful things positive deviants do and eliminate those that are true but useless--TBUs. TBUs are those factors which may truly contribute to the success of the individual using them, but, because they are not accessible to all and therefore cannot be practiced by everyone in the group, they are useless in solving the problem.

For example, if a teacher was clearly more successful than her colleagues in teaching reading, but we discovered that she had only 10 students in her class while other teachers in her school had 20 students, we would call that finding a TBU. While it is probably true that a smaller class size contributed to this teacher's success, it is useless at the moment because we can't tell the other teachers to get rid of half their students. Likewise, if we discovered that the positive deviant teacher was more creative, we would consider that characteristic a TBU. While her creativity might truly contribute to the outcome, we can't wave a magic wand and tell other teachers to be more creative. By focusing on what the creative positive deviant teacher actually does, we are likely to discover concrete strategies that are accessible to all teachers regardless of their personal creativity. The utility of the TBU concept is that it provides a sieve through which we pass the uncommon qualities of positive deviants to make certain that we've identified only those which can be practiced by everyone.

A Positive Deviance Inquiry

JSD: What does a Positive Deviance Inquiry look like from beginning to end?

Sternin: It has just four steps--define, determine, discover, and design. The group begins its work by defining the problem and describing what success would look like--which is the inverse of the problem statement. Next, the group determines whether there are individuals who have already achieved success. If there are such people, they are the positive deviants. Next, the group discovers the uncommon but demonstrably successful behaviors and practices used by the positive deviants to solve the problem. And finally, the group designs an intervention which enables its members to practice those demonstrably successful but uncommonly applied practices. The process is beautifully simple because its strength lies in the solutions that are discovered and owned by people in the community.

Here's what these steps look like in practice. I worked with a group of AmeriCorps and Volunteers in Service to America (VISTA) volunteers who were part of a Save the Children program in Appalachia. The number of students with learning disabilities in the community was extraordinarily high. The slice of this problem that the group--which included some parents--wanted to look at first was why some learning-disabled kids did much better in school than others.

One of the things the volunteers discovered in the Positive Deviance Inquiry was that the parents of the more successful students would come to the school, meet with the teachers at the beginning of the year, and enlist their support in dealing with problems before they became serious. The other parents in this community did not do that; they had contact with teachers only when there was a problem. They said they felt embarrassed and didn't know how to approach a teacher.

As a result, the positive deviant parents offered weekly workshops in which they used role-playing to teach shy, inarticulate parents how to talk to teachers. The positive deviance method can be contrasted with knowledge-transfer approaches, in which an expert tells parents, "You should go to teachers at the beginning of the school year to talk to them." That doesn't work.

Look within for resources

JSD: The physical presence of the positive deviant in the affected community seems to be a significant aspect of this approach. It can be contrasted with an approach in which researchers study the practices of successful teachers in various subject areas and the qualities of schools that produce more learning than other schools with the same type of student population. While those teachers and schools might be thought of as positive deviants, what researchers learn usually doesn't transfer well to other settings.

Sternin: The Positive Deviance Approach requires that community members find the positive deviants within their own community. The community is self-defined, and its members always share the same resource base. If teachers in a school feel that the school is a community, then it's a community. If 10 schools in a district say that collectively they are a community, then they are a community. If five of them, however, say that they are a community because of particular characteristics they have in common, then they are the community. That's important, and it's a critical distinction between a best-practices approach and positive deviance.

A different approach

JSD: Here's another strategy that at first might appear to be a Positive Deviance Approach to improvement. A principal who's heard about positive deviance invites a teacher who has consistently produced higher reading test scores to tell her peers how she does it. What often occurs, however, is that the faculty resists, and the teacher the principal has identified may even feel shunned by her colleagues.

Sternin: Let me contrast what you've described with the Positive Deviance Approach. We would begin by defining the problem. In this case it sounds like it is low reading performance by students. A facilitator might say to the principal and teachers, "I was asked to come by today because the principal told me that test scores indicate 65% of students in this school are not learning to read very well. To begin, I'd like to know if you agree this is a problem."

Usually, teachers agree there is a problem. If they don't, that's the end of the inquiry. If teachers agree, they would be asked to define a successful resolution of the problem. Let's say they agreed that success would be 90% of students learning to read at a particular level. Next, without considering the names of teachers, we would examine test scores by classroom to see which ones met or at least closely approximated that outcome. These are the positive deviants. We would then look more carefully at those situations to make certain those teachers do not have access to special resources such as aides or smaller class sizes.

Teachers would then list the broad areas they want to examine more carefully and determine the norms of current practice in their classrooms in each of these categories. They would develop a checklist for observation and write interview questions based on the categories they identified. The group decides who will observe and do interviews and when they will be completed. Based on that information, the group determines what the positive deviant teachers are doing that is different from normative practice. Next, the group identifies five or six strategies that are accessible to everyone and decides how group members will practice and learn these strategies. It is important to underscore that teachers design the intervention that enables them to practice the new behaviors.

in the worst circumstances

JSD: Many schools, particularly low-performing ones, suffer the effects of a number of long-standing societal problems such as poverty and racism. In addition, adults in these schools may not be accustomed to working together and are often angry and distrustful. I'm curious about the success of a Positive Deviance Inquiry in the face of such obstacles.

Sternin: The Positive Deviance Approach has proven its worth in the most difficult situations, not the best. In Argentina, for example, where our work is still in its early stages, the World Bank asked us to use positive deviance in a very poor, remote part of the country. The minister of education and the head of a group that represents teachers were very concerned about the retention of students in schools. Across a district of 120 or so schools, 40% of the students did not graduate from primary school, but at the same time, some schools were retaining 80% or 90% of their students. Because of the economic crisis in the country, teachers were incredibly demoralized and very angry because they hadn't been paid in six months and because they felt like parents didn't care about their kids.

To see if the Positive Deviance Approach made sense in this setting, I scheduled meetings with teachers, administrators, and parents. At the first meeting, 25 teachers and administrators sat with their arms folded across their chests indicating, "OK, Mac, do it to us. We've been called away from school for this workshop, we're really angry, and we don't want to be here."

I spent the first 40 minutes of the meeting talking about positive deviance and how it's been used in other settings. At that point a woman said, "Look, Argentina is not Vietnam or Burma. You don't understand that we haven't been paid in a long time and that we have these incredible problems in our schools. Maybe positive deviance worked someplace else, but it can't work here." Then she started listing all the things teachers were really angry about. I told her that I heard what she was saying and that I was sure the things she'd mentioned were important. I also explained that the positive deviance process provides a way to not become immobilized by the magnitude of these issues. I asked the group to trust me for another hour if I promised I would get back to the list of problems raised by the teacher. Early in the day I reminded participants that the next day they were supposed to bring 20 parents with them. Teachers immediately responded that they couldn't get that many because parents weren't interested, but that they would try to get two or three to attend.

The day turned out very well because we looked at what was going right rather than wrong, and because we weren't trying to find fault or lay blame. At the end of the day the head of the teachers group said it had been a long time since he had seen teachers leave a meeting feeling positive rather than angry. The following day teachers brought with them 25 parents, which meant a drive of several hours because of the distances involved. Teachers were so excited about the potential of finding solutions within the context of their demoralized world that they recruited those parents and drove them all the way to the meeting.

Honor participants

JSD: As I listen to you describe this process, I am struck by how honoring and appreciative it is of the people who engage in it.

Sternin: That's it. As hostile as the teachers were the day before, the parents that day were quiet and shy. But we made it clear that we wanted to learn from them. They told us that this was the first time anyone had asked them for their ideas. By the end of the day, though, they truly believed they were experts within the context of this problem and their community. Positive deviance is a very empowering approach, but it's one that individuals with lots of degrees on their walls may find difficult to implement. Positive deviance inquires into what's working and how it can be built upon to solve very difficult problems. It requires that experts relinquish their power and believe that solutions already reside within the system. Our role is to help people discover their answers.

References

Dorsey, D. (2000, December). Positive deviant. Fast Company, 41, 284-292.

Haycock, K. & Huang, S. (2001). Are today's high school graduates ready? Thinking K-16, 5(1), 3-17.

Pascale, R.T., Millemann, M., & Gioja, L. (2000). Surfing the edge of chaos: The laws of nature and the new laws of business. New York: Crown Publishing Group.

Temes, P. (2001, April 4). The end of school reform. Education Week, 20(29), 36.

Wagner, T. (2001, January). Leadership for learning: An action theory of school change. Phi Delta Kappan, 82(5), 378-383.

JERRY STERNIN

Position: Jerry Sternin is a visiting scholar at Tufts University and an international consultant. He also is involved in Positive Deviance projects for the World Bank in Argentina and for USAID-funded development organizations in Indonesia, and he recently presented the Positive Deviance approach to corporate and social entrepreneurs at the World Economic Forum in Geneva.

Education: Sternin has a bachelor of science degree in speech therapy from the State University of New York at Geneseo and a master of arts degree in Asian studies from Harvard University.

Professional history: He has 24 years' experience in developing countries, including eight years with the Peace Corps in the Philippines, Nepal, Mauritania, and Rwanda and 16 years as a Save the Children director in Vietnam, Bangladesh, Egypt, the Philippines, and, most recently, Myanmar.

Publications: Sternin published an article, "The Power of Positive Deviance," in the Harvard Business Review in January 2000, and in December 2000, he was featured in an article on Positive Deviance in Fast Company magazine. The articles brought many queries from the corporate world on the use of Positive Deviance and have helped to broaden the Positive Deviance audience.

Accomplishments: Sternin has served as an assistant dean and adviser to students at the Harvard Business School. He also received a writing grant from the Rockefeller Foundation and is currently the recipient of a Ford Foundation grant to amplify Positive Deviance.

Personal: Sternin's wife, Monique, has been his partner in developing the Positive Deviance approach. They have one son, Sam, who works for Save the Children in Vietnam.

To continue this conversation with Jerry Sternin, contact him at jerry_sternin@hotmail.com.

The 6 Ds of the Positive Deviance Approach

Define

What is the problem and what are the perceived causes and the related community behavioral norms?
What would a successful solution/outcome look like (described as a behavioral or status outcome)?

Determine

Are there any individuals or entities in the community who already exhibit desired behavior or status (identification of positive deviants)?

Discover

What are the unique practices or behaviors that enable positive deviants to outperform or find better solutions to problems than others in their community?

Design

Design and implement an intervention that enables others in the community to experience and practice new behaviors (focus on doing rather than transfer of knowledge).

Discern

What is the effectiveness of the intervention (determined by ongoing monitoring and evaluation)?

Disseminate

Make the intervention accessible to a wider constituency (replication/scaling up).

NSDC Book Published

With funding from the Kellogg Foundation, NSDC began in 2003 to seek ways that positive deviance can be applied in education. Joan Richardson, NSDC's director of publications and JSD executive editor, visited six schools and districts the organization identified as being positive deviants. These schools and districts have achieved above-average results with students although they have access to the same resources as other schools and districts in their areas. In addition to standing out in their communities, these schools and districts have identified good practices internally and ensure that these practices are shared widely throughout the school or district, enabling all teachers to perform at higher levels.

NSDC will publish From the Inside Out: Learning from the Positive Deviance in Your Organization in 2004, and excerpts will be published in Results. Order your copy online at store.nsdc.org.

Idea is tested at a time of dire need

I first became aware of the Positive Deviance Approach to problem solving in a December 2000 Fast Company article that described Jerry Sternin's work in Vietnam for Save the Children. A few months later, in an interview with JSD, Robert Quinn used the term and recommended I read Surfing the Edge of Chaos (Crown Publishing Group, 2000), by Richard Tanner Pascale and several colleagues, for a more complete description.

Pascale tells how malnutrition was a serious problem in Vietnam following the war, a problem not unlike those faced by other developing countries, problems typically viewed as unsolvable because of systemic properties (poverty, low levels of education, lack of access to clean water and sanitation). Attempts to address these larger problems or provide a massive infusion of supplemental food had provided little long-term benefit.

Within that context, Save the Children decided in the early 1990s to try a living systems approach known as positive deviance.

"Positive deviance," Pascale writes (pp. 176-177), "does not impose a nutritional solution. Rather, this model relies on ... scaling up a solution that is already working in the community. ... The design was aimed to discover what was already working against all odds, rather than engineering a solution based on an external formula." Each community examined its conventional wisdom regarding nutrition and health care, Pascale notes, and conducted an inquiry to uncover unconventional nutritional practices advantageous to children's health, then made those practices accessible to everyone.

Pascale and his colleagues report that within six months, more than two-thirds of the children had gained weight. Within two years, 85% were no longer clinically malnourished. The Positive Deviance Approach "... was scaled up consistent with its philosophy of discovering unique positive deviant solutions in each area - an approach that is very different from a socially engineered 'best-practices' rollout. ... Essential to this approach is first, respect for, and second, alliance with the intelligence and capacities residing within the village. This model can be applied to other kinds of change . ... The wisdom to solve problems exists and needs to be discovered within each and every community," they conclude (pp. 177-179).

In the next few months, I came across a number of school reform-related items that in my mind seemed linked to the Positive Deviance Approach. For instance:

Peter Temes (2001), president of the Great Books Foundation, recounted the story of a recently-retired superintendent who told him, "The very best thing you can do for a superintendent is not to give him more money, more buildings, or a better contract. Instead, give him a tool to make his average teachers just a little bit better, and you'll see a vastly greater impact across the district than any model school or blue-ribbon program will ever bring" (p. 36).
Kati Haycock and Sandra Huang (2001) cited a Boston study of teacher effects that they describe as fairly typical. The study revealed that in just one academic year, the top third of teachers produced as much as six times the learning growth of the bottom third.
Tony Wagner (2001) concluded: "The task of the leader is not to tell teachers what (best practices) are but to create opportunities for educators to discover them for themselves. ... Effective leaders give teams of experienced teachers--the building leaders--time to visit successful schools and to discuss what they've learned with colleagues. Teachers need to see models of much more successful classrooms in order to believe that all students can succeed" (p. 382).

Average teachers doing a little bit better. The practices of our most successful teachers being spread within their schools. Educators discovering best practices for themselves within school cultures that nurture such discovery and support their amplification throughout the school. In a nutshell, that is my sense of the Positive Deviance Approach to change.
- Dennis Sparks

About the Author

DENNIS SPARKS is executive director of the National Staff Development Council.