Applying systems thinking to school leadership Part 2

Getting technical with Donella Matthews

In Part 1 of this series I looked at the idea that ‘a system is perfectly designed to get the results it gets’¹ and explored some of the more predictable reasons systems go awry. This post will address the idea of systems engineering in a bit more detail and spotlight the work of environmental scientist Donella Meadows.

---

In Thinking in Systems: A Primer, Meadows points out that systems are everywhere but are often hidden from view. From ecosystems and organisations to economies and your morning routine, systems are interconnected sets of elements working toward a purpose. Because we don’t usually perceive the world around us as being made of up systems we’re prone to falling into various cognitive traps which cause us to blame or credit individuals for events rather thann looking for their true causes.

No one deliberately creates problems. The endemic problems found in schools - poor reading, unruly behaviour, teaching to the test, data misuse, the advantage gap, teacher workload - are not caused by bad actors. They are baked in responses to the structures and systems schools perpetuate and are part of. It doesn’t have to be this way. A few schools have managed to escape some of the endemic traps into which most fall, but working out how and where to adjust systems is hard. Understanding that systems create their own results, regardless of the intentions of whoever might have built them provides a fundamental insight into understanding human behaviour.

In this post we’ll look at the various elements of system interconnectedness and explore how they can be applied to the complicated business of running schools. but before we embark, a word of caution: all the ideas that follow rely on models of the world, not accurate descriptions of reality. As mathematician George Box famously said, “All models are wrong but some are useful.” Meadows is at pains to point out that no matter how practised you become at seeing the world as sets of interconnected systems, you’ll still be surprised by reality because no model can account for every variable.

Stocks & flows

Systems manage ‘stocks’ (anything that can be measured at a point in time - the ‘stuff’ in a system that accumulates or depletes over time) which are affected by ‘flows’ (anything that increases or decreases a stock - whatever’s going into or out of a system). If we think of a bathtub as a system, the stock is the water leve and the flow is represented by the taps and drain. Turn on a tap and the stock rises; pull the plug and it drains away.

If we apply this simple model to, say, a school’s behaviour management system, the stock could be the overall level of positive student behaviour, the quality of school culture and the amount of trust or respect within the community. The flows are influences on these stocks:

This model can help us to think systemically, rather than look for someone to blame, and encourage us to ask, What are we doing every day that fills or drains the behaviour bucket? Asking this question helps to think in terms of an interconnected system rather than simple cause and effect. For instance, are we training and supporting staff to respond to students’ behaviour predictably and using the same language? Do students feel heard and valued? Are the systems we’ve designed easy for staff to use? Do families understand our rules and expectations? Are there particular times in the day or week or term when behaviour is better or worse?

Both inflows and outflows can take a while to affect behaviour in a school, but over time the stock will raise or lower - that is to say, behaviour will get better or worse - depending on how flows are managed. We can’t just react to negative events (outflows) we also have to intentionally increase inflows to build a positive culture. Or, to return to the bathtub analogy, you have to put in the plug and turn on the tap for the water level to rise. But, obviously, the culture in a school is a much more complex system than a bathtub. In the first part of this series I discussed how perverse incentives can be caused by having too rigid a control of inflows and lethal mutations by having too laissez faire an approach. To stay on top of the balancing act required to keep systems behaving optimally, it helps to think in terms of feedback loops.