The Cynefin Framework and (the Complexity of) Classroom Instruction

by ajc

Classroom instruction is complex but do we treat it as such? Is “sensing” a priority of teacher education? How would an instructor who waits for “patterns to emerge” be viewed by their supervisor? As laid back? Aloof? And does outcome-based education (unintentionally) result in educators treating complex situations as complicated, or worse yet, simple in nature? These questions find their origins in the work of David Snowden and Mary Boone, as they apply the principles of Snowden’s Cynefin Framework to leadership in the 2007 Harvard Business Review article A Leader’s Framework for Decision Making.

Rather than summarize Snowden & Boone’s application textually, I’ve generated a graphical representation. Clicking on this smaller image will produce the full-sized document.


Initially, my interest in this article was peaked by a blog posting of David Clark in his series of writings aimed at describing how the tenets of Agile software development can be applied as a system of instructional design. The HBR article had the added advantage of using this framework as a lens from which to view leadership, a topic of focus for me now as I begin supervising staff at my school. However, its application to the classroom teacher is intriguing.

Aren’t teachers really classroom leaders, or leaders within a complex educational / instructional setting? David Williamson Shaffer, in his 2004 TCR article Pedagogical Praxis: The Professions as Models for Postindustrial Education states the following.

We know a great deal about some of the epistemological and pedagogical underpinnings of compelling learning environments. However, orchestrating these elements into a coherent whole remains a challenge (p. 1417).

Shaffer suggests that professions have unique epistemologies, or “ways of knowing”, and thus have the potential to be used as models for instruction. Daniel Willingham, in his new book Why Don’t Students Like School?, alludes to the complexity of the educational setting as well as he explains the disconnect between controlled experimental studies and the classroom.

But mental processes are not isolated in the classroom. They all operate simultaneously, and they often interact in difficult-to-predict ways (Kindle edition, Location 126).

Marzano et al., in their work Classroom Instruction that Works: Research-Based Strategies for Increasing Student Achievement discuss nine categories of instructional strategies.

  • Identifying Similarities and Differences
  • Summarizing and Note Taking
  • Reinforcing Effort and Providing Recognition
  • Homework and Practice
  • Nonlinguistic Representations
  • Cooperative Learning
  • Setting Objectives and Providing Feedback
  • Generating and Testing Hypotheses
  • Cues, Questions, and Advance Organizers

Throughout this work, the presented strategies are also (necessarily) related to content and instructional situation. Doesn’t this make instructional choices more complex? If we agree with Willingham, who differs from Marzana et al. in this regard, there’s another layer of complexity that must be superimposed onto the extant relationships between content and strategy, topic specific ability. He contends that we must present problems to learners that fall within a specific range, not too tough (which causes despair), or too easy (no satisfaction, nor release of dopamine), but just right. (Goldilocks analogies are welcome.)

Thus, we will seek out opportunities to think, but we are selective in doing so; we choose problems that pose some challenge but that seem likely to be solvable, because these are the problems that lead to feelings of pleasure and satisfaction (Kindle edition, Location 400).

At this point, I’ve identified several variables that must be considered by a teacher as they teach.

  • What needs to be taught
  • What has been taught
  • What is to be taught in the future
  • Individual abilities of students with regards to content
  • Individual abilities / preferences of students related to instructional strategies
  • Situation / relation between content and potential strategies (what “fits” best)

But there is another, most important factor, life outside of the classroom. What happens beyond the classroom walls, in other classes, and more significantly outside of school, affects each learner. The combination of these variables supports the idea that classrooms should be classified as “complex” with the Cynefin Framework. If we review the traits of “Complex” systems, it is clear that often times there is “no right answer” in terms of instructional choices, that classrooms are “systems in constant flux”, and that the “ability to understand” (from the teacher’s perspective) comes after class has been dismissed.

The ability to work within this complex system (the classroom) is typically part of the teacher observation process. For example, Domain 3e in Charlotte Danielson’s Framework for Teaching is entitled Demonstrating Flexibility and Responsiveness, and includes the elements Lesson adjustment, Response to students, and Persistence. However, outcomes are the focus of this work, not the manner in which those outcomes are reached. Admittedly, factors such as rapport and teacher-student interactions are also part of these assessments, and these might be considered constituents of “probing”. However, the emergence / identification of patterns is not addressed in any meaningful way.

Is there a need, then, to construct a formalized framework / structure for “probing” and “sensing (for emergent patterns)” specific to the classroom? Is this something teachers need to know how to do? Would familiarity with the tenets of the Cynefin framework lead to more effective instruction through appropriate responses to the different categories of complexity manifest within that setting?

Comments and/or critiques are welcomed.