There are four types of problems: Simple, Complicated, Complex, Chaotic. And there are four recommended approaches to solving them.
In this posting, we will borrow from David Snowden’s Cynefin Framework, and map the application of the principles both to the problems confronting education reform and also to the practice of problem based learning in schools.
Simple problems are those with known solutions, where there are predictable and repeatable cause-and-effect relations. These problems lend themselves to algorithms and best practice analysis, “whenever this happens, do this.” You sense the situation that exists, you categorize what type of problem it is, and then you respond according to the best practice. The shorthand for this type of problem solving sense, categorize, respond. Bureaucracies handle these types of problems very well.
English: Cynefin framework with all five domains labeled (Photo credit: Wikipedia)
Complicated problems generally require more analysis, but the results can still be predictable. It is more likely that there is no single solution to the problem, there may be many solutions that could work depending on the skills and disposition of those involved. Often, no two situations are exactly the same, so one is dependent on extrapolating from past experience and devising custom solutions. Because there is no one “best practice” experts determine “good practice”, solutions that should work based on the skill sets of those involved. In these problems, you sense or get data about the situation, you apply your expertise to analyzing the information to determine a likely course of action, and then you respond based on a plan that came out of your analysis. Professionals, rather than bureaucracies, are needed for these situations because they can devise customized plans based on their knowledge and experience.
Complex problems have no clear solution(s) except in hindsight; they are unpredictable. The value of extensive planning is diminished, because it will not yield a “good practice”. Algorithms are not useful, because there are no clear cause and effect rules. No one can predict which course of action will work, so one would want to try many different actions, with relatively little vested in any one (safe-to-fail actions or probes), and then reinforce those that are working and diminish those that are not. The approach to these problems would be to try different actions or probe, figure out which ones are and are not working, and then respond accordingly to amplify what’s working and curtail what is not. From these probes or actions, a practice will emerge that, in hindsight, will seem the one that would obviously work. The shorthand for this approach is probe, sense, respond.
In a chaotic problem, things are seemingly spinning out of control. There are so many factors happening at once that one cannot get a grasp on the relationships between them, and the situation itself is probably changing rapidly. This is an unwieldy situation, where no solutions are evident. A chaotic situation might inspire creativity, or it could demand some immediate triage actions to try to bring it under control, sense whether more triage is needed or whether solutions can emerge, and then to respond. The shorthand for this approach is act, sense, respond.
How does the Cynefin framework relate to education policy?
Much of the time we are sitting inside a problem, and select the problem solving mode that we are most familiar with.
A bureaucrat might look at education improvement as a simple or cause and effect problem and say, “the problem with education is X, and that problem just needs Y.” For example, “The problem with education is that we have no national standards, so we just need a common core of standards and our educational system will improve.” Or “The problem with education is that we have too many bad teachers, so we need to measure teachers and hold them accountable and then our educational system will improve.”
An educational professional might look at education improvement as a complicated issue, one where there are known approaches, but solutions need to be individually crafted based on the state, district, school, or student. For example, “if we assess each student and determine his or her strengths and weaknesses, then we can individualize instruction for that student so each student will learn at his or her potential.”
An entrepreneur might look at education improvement as a complex issue, one where we don’t know what will work, but one where a solution would emerge through trial and error. For example, “Let’s start a bunch of charter schools and allow them flexibility so that we can then scale what’s working.”
A CEO might look at education improvement as crisis management, “Let’s first make sure everyone has basic reading, writing, and arithmetic covered. We will give each teacher prescribed lesson instructions and make sure the students get lots of practice, and then we can start dealing with the other problems and solutions.”
We need to recognize these biases as we look to improve the different facets of the education system:
- Are we in a chaotic crisis?
- Are we looking for some emergent solution?
- Is this something that has multiple solutions and where implementation will require customization and expertise?
- Is there one best practice that will work?
How does this framework apply to the classroom?
We already give students tons of simple problems, problems that have one solution and that primarily require them to look at the problem, determine the formula to use, and then apply the formula in order to get a good grade.
Some students are also being given the opportunity to solve complicated problems. These problems do not have one right answer, students learn to get information, analyze and plan, and then operate against their plan to produce an end product.
There is a nascent movement around “engineering design” that presents students with problems that require an iterative process of sensing, probing, sensing, and reacting.
And chaotic problems? Do you know cases where students are confronted with unwieldy situations that are spinning out of control? Perhaps we should use Washington as a model for that.