Tuesday, November 15, 2005

Defining the System to be Improved

Piecemeal change to improve schooling inside a school district is an approach that at its worst does more harm than good and at its best is limited to creating pockets of “good” within school districts. When it comes to improving schooling in a district, however, creating pockets of good isn’t good enough. Whole school systems need to be improved.

To transform an entire school system, change leaders in that system must know what a system is and how it functions and they must be skillful in using a set of systems thinking tools. This blog introduces you to both of these competency sets.

The concept of “systemic change” provides the context for the kind of change leadership described in this blog. Because the term has different meanings, I need to clarify exactly which meaning I will be using. Squire and Reigeluth (2000) identify four distinct meanings of the term:

Statewide policy systemic change. This meaning focuses on statewide changes in tests, curricular guidelines, teacher-certification requirements, textbook adoptions, funding policies, and so forth. These changes are supposed to be coordinated to support one another (Smith & O’Day, 1990). This meaning is frequently used by policymakers when they talk about systemic change.

Districtwide systemic change. Educators subscribing to this meaning see systemic change as any change, including new programming, intended to spread across an entire school district. This is the meaning often held by preK-12th grade educators.

Schoolwide systemic change. Using this meaning, educators see systemic change happening inside single school buildings and it typically involves “…a deeper (re)thinking of the purposes of schooling and the goals of education” (Squire & Reigeluth, 2000, p. 144). This is the meaning that seems to inform the work of such groups as the New American Schools, Inc. and the Coalition of Essential Schools.

Ecological systemic change. This meaning sees systems as rich networks of interrelationships and interdependencies within the system and between the system and its “systemic environment” (the larger system of which it is a part, its peer systems within that larger system, and other systems with which it interacts outside of its larger system). This perspective recognizes that a significant change in one part of a system requires changes in other parts of the system. It also recognizes the need for changes in three interconnected aspects of a system: its core and supporting work processes, its internal social architecture, and its relationships with its environment (Duffy, Rogerson, & Blick, 2000). This view of systemic change subsumes the other three meanings, and it is how “systems thinkers” view systemic change (e.g., Ackoff, 1981; Banathy, 1996; Checkland, 1984; Emery & Purser, 1996; Senge, 1990). This is the definition that I endorse.

Russell Ackoff (1999, pp 6-8) adds depth and breadth to our understanding of organizations as systems. He says a system is a whole entity consisting of several parts with the following properties, which were edited to fit school systems:
  • The whole school system has one or more defining properties or functions; for example, the defining function of a school district is to educate children and teenagers.
  • Each piece of a school system can affect the behavior or properties of the whole; for example, a couple of low performing schools in a district can drag a whole district into low performing status.
  • There is a subset of school system components that are essential for carrying out the main purpose of the whole district but they cannot, by themselves, fulfill the main purpose of a school system; e.g., teachers and classrooms in a single school building are essential elements of a school system and they are necessary for helping a school system fulfill its core purpose, but these classrooms and schools cannot and never will be able to do what the whole school system does.
  • There is also a subset of components that are nonessential for fulfilling a school system’s main purpose, but are necessary for other minor purposes; e.g., school public relations, secretarial work, and pupil personnel services.
  • A school system depends on its environment for the importation of “energy” (i.e., human, technical and financial resources); therefore it is an “open system.” A school district’s external task environment (i.e., the environment it interacts with on a daily basis) consists of individuals and groups identified as customers, critics, competitors, suppliers, and stakeholders.
  • The way in which an essential element of a school system affects the whole system depends on its interaction with at least one other element; e.g., the effect a single school’s performance has on the whole district depends on the interaction that the school has with at least one other school in the system.
  • The effect that any subset of a school system has on the whole system depends on the behavior of at least one other subset of elements; for example, let’s say that a school district is organized preKindergarten-12th grade. This means the core work process for that district is 13 steps long (pre-K-12th grade).
    Now, let’s say that district leaders are concerned about the performance of their high school (which is a subset of the whole system). This high school contains grades 9-12. It would be a mistake to focus improvement efforts only on that high school because its performance is affected by at least two other subsets of schools (i.e., the elementary and middle schools that “feed” into the high school).
    Since all essential elements of a school system interact, it would be wise to examine and determine how the elementary and middle schools are affecting the performance of the high schools. Focusing improvement only on the high schools would be a non-systemic and, therefore, piecemeal approach to improvement.
  • A school system is a whole entity that cannot be divided into individual components without losing its essential properties or functions. For example, the dominant approach to improving schooling in the United States is called school-based or site-based improvement. This approach divides a school system into its aggregate parts; i.e., individual schools. Then, it is assumed that improving these individual schools will somehow improve the whole system. When attempts are made to improve a school system in this way—by disaggregating it into its individual schools—its effectiveness as a system deteriorates rapidly.
  • Because a school system derives its effectiveness from the interaction of its elements rather than from what the elements do independent of the system, when efforts are taken to improve the individual elements as if they were not part of a whole system (as in school-based improvement), the performance of the whole system, according to Ackoff (p. 9), deteriorates and the system involved will be significantly weakened.

    Strategies for Defining the System to Be Improved

    Some authors (Fullan, date unknown) are suggesting that systemic change in education must occur within a mega-system defined by three levels: the district system, the community system and the state department of education system. This tri-level solution, as Fullan calls it, seems reasonable and intuitively correct. I agree that all three systems must be engaged in any effort to transform a school system. Although I agree with this tri-level concept, I would like to tweak it using principles from systems theory, in particular principles from Ackoff (above) and Merrelyn Emery (described below).

    Let’s examine the tri-level solution keeping Ackoff’s systems principles (above) in mind.

    Each school district, community and state department of education interact with each other and to some degree depend on each other, but each also functions as an intact, self-managing system. When the three are combined into a mega-system for the purpose of systemic change, the complexity of that system countervails any effort to improve that mega-system.

    As a case in point, consider what a mega-system of education would look like in the state of Pennsylvania. That state has 99 school districts, each inside a community, thereby adding 99 additional systems to the mega-system. Finally, there is one state department of education. The total number of self-functioning, relatively autonomous sub-systems in that mega-system would therefore add up to 199 complex sub-systems (99 communities, 99 school districts and 1 state department of education), each with its own special characteristics.

    How would you go about changing that tri-level mega-system? Who would lead that transformation? How would the special characteristics, needs, and demands of each of the 199 sub-systems be addressed? How long would it take to transform a mega-system of this size if, indeed, it could be changed at all?

    I think it would be impossible to change a mega-system of that size and complexity.

    Merrelyn Emery (Emery & Purser, 1996) gave us a different strategy for identifying “the system to be improved.” She said that we should draw a circle around every unit, department, person, and so on that must collaborate to deliver a complete service or product. To identify a unit of change that would increase the likelihood of improving schooling, that circle would go around everything that we call a school system. Everything inside the circle—the entire school system—becomes the unit of change. Everything outside the circle is part of a school system’s external environment.

    The idea of involving three system levels in any effort to improve a school district is correct. How to involve those three systems is where there is disagreement. Instead of creating a single tri-level mega-system to involve a school district’s community and state department of education, if we used Emery’s principle for defining the system to be improved those two systems would be considered part of a school district’s external environment. Then, change leaders could follow three paths simultaneously that would lead to school district transformation.

    Three Paths Toward Improvement (Duffy, 2003, 2004)

    Principles of whole-system change suggest that three paths must be followed to change and entire system (where a system is defined using Merrelyn Emery’s advice). The three paths are: Path 1: improve a district’s relationships with its external environment; Path 2: improve a district’s core and supporting work processes; and Path 3: improve a district’s internal social environment.

    Using this three path approach (rather than a tri-level approach) change leaders in school systems would work with their colleagues to improve the relationship they have with their community and with their state department of education. Using this three-path approach in Pennsylvania, for example, each of the 99 school systems would be working to improve their unique relationship with their own community and each would be working to improve their unique relationship with their state department of education. Thus, systemwide change would be occurring with each school system by engaging that district in tri-level relationship (each district with its community and with its state department of education) thereby making the prospects of successful systemic change more likely than if this same kind of change process was attempted in a mega-system.

    It is clear that the community system and state department of education system must be part of any district’s effort to engage in whole-system change. Blending the three systems together to create a single mega-system creates an extraordinarily complex mega-system that may become notoriously impossible to change and would create significant resistance to future change.


Ackoff, R. L. (1981). Creating the corporate future. New York: John Wiley & Sons.

Ackoff, R. L. (1999). Re-creating the corporation: A design of organizations for the 21st century. New York: Oxford University Press.

Banathy, B. H. (1996). Designing social systems in a changing world. New York: Plenum Press.

Checkland, P. (1984). Systems thinking, systems practice (Reprinted with corrections February, 1984 ed.). Chichester Sussex. New York: Wiley & Sons.

Duffy, F. M. (2004). Courage, passion and vision: A guide to leading systemic school improvement. Lanham, MD: ScarecrowEducation and the American Association for School Administrators.

Duffy, F. M. (2003). Step-Up-T0-Excelllence: An innovative approach to managing and rewarding performance in school systems. Lanham, MD: ScarecrowEducation.

Duffy, F. M., Rogerson, L. G., & Blick, C. (2000). Redesigning America’s schools: A systems approach to improvement. Norwood, MA: Christopher-Gordon Publishers.

Emery, M. & Purser, R. E. (1996). The Search Conference: A powerful method for planning organizational change and community action. San Francisco: Jossey-Bass.

Fullan, M. (date unknown). The tri-level solution: School/district/state synergy. Retrieved on May 7, 2005 at http://www.saee.ca/analyst/C_023.1_BII_LON.php.

Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization (1st ed.). New York: Doubleday.

Smith, M. S., & O’Day, J. (1990). Systemic school reform. In S. Fuhrman & B. Malen (Eds.). The politics of curriculum and testing (pp. 233-267). Philadelphia: Falmer Press.

Squire, K. D., & Reigeluth, C. M. (2000). The many faces of systemic change. Educational Horizons, 78(3), 145-154.

No comments: