Desirable difficulties, teacher development and creating a climate of continuous improvement

There is no such thing as the perfect teacher: we can all improve. Most of us have areas of strength and weakness. We will be aware of some of them and others will be hiding in our blind spot. So schools, rightly, invest a large portion of time and resources on developing teachers.

The evidence suggests most teachers improve over their first 5 years and then plateau. The reason for this, and so many other issues in schools, is complicated, but the environment the teacher works in plays a large role. So I have been thinking about how best to create a climate of continuous improvement in my department.

Fig 5 in the link above

How do schools often go about improving teaching?

In my experience the most common way schools go about improving teaching is to publish a set of agreed criteria for ‘good’ teaching and then quality assure their staff against the criteria. The staff that are found not to be meeting these expectations then are placed within a system of increasing levels of formal support. These packages start by working with a head of department, meeting once a week. They then go through a series of steps over a few terms and end with formal capability proceedings. This ensures all staff meet a minimum expectation. This obviously works and now all teachers are good and everything is awesome.

Or not.

This approach has a number of drawbacks:

  1. It creates a stigma where support and development is due to apparent failure.
  2. Staff that just meet the criteria are not supported as much as they might need to be.
  3. The strong teachers get completely ignored.

Most importantly, any high stakes criteria leads to gaming of the system. according to Donald T Campbell:

The more a quantitative metric is visible and used to make important decisions, the more it will be gamed—which will distort and corrupt the exact processes it was meant to monitor.

Adaptation of Campbell’s law

I’m sure we have all fallen foul of this (*cough* triple marking *cough*), and the concept of metrics and their unintended consequence is a blog unto itself. For now I will just recommend you read The Tyranny of Metrics by Jerry Z Muller. It’s essential for any level of school leadership.

The formal support packages commonly used in schools create a high-stakes, short-term, intensive environment with two or three large areas of focus. The degree of difficulty for a struggling teacher is probably too high to be effective if the issue is one of the teacher’s skill. Think about being this teacher – maybe you have been there yourself? They are losing time to have the meetings, have a huge feeling of failure and are under a large amount of pressure to hit multiple targets in a time frame of a few weeks.*

What about the rest of us?

To solve drawbacks 2 and 3 the school will have a rolling CPD program. Normally this is a mix of voluntary and compulsory sessions. This will probably include Inset Day training, learning communities or peer-to-peer coaching. This in turn definitely works and all staff should improve, right? Well obviously not, as any teacher who has been to CPD can attest. You might gain one or two nuggets but large-scale change is rare.

Two factors that we know are hugely important in student learning are often ignored in teacher development:

  1. The value of practice to develop mastery
  2. The power of desirable difficulties

This blog will focus more on the concept of desirable difficulties and its implications for teacher CPD.

What are desirable difficulties?

Robert A. Bjork first coined the term ‘desirable difficulties’ in 1994 to describe the ‘Goldilocks zone’ of optimum learning. Tasks should be hard enough to be effective but easy enough to motivate learners to try. Put simply, we often make our decisions by the rule:

Is the juice worth the squeeze ?

Not this Bjork**

As teachers we spend a lot of our time considering the level of desirable difficulties in our learning activities. I have written previously about ‘boiling the frog’. But as leaders I wonder how often we consider the desirable difficulties of our teacher improvement strategies?

(At this point it is worth mentioning that if it is merely an effort problem then these systems will probably be appropriate, just like a novice student cannot answer synoptic questions but an experienced student can. These cases are most likely rare, but they do occur and a valuable action of teaching and learning quality assurance is to identify these situations.)

For most teachers the best way is to deliver improvement is to work on small areas incrementally. Teachers agree an area to develop via discussion with the observer. This area receives deliberate practice in their lessons over time. The same observer visits further lessons to provide guidance on their progress. Once this area has been developed then they will agree on the next target. In this way every teacher has one priority they are working on, but only one. In my opinion this approach is akin to diffusion *** (or an ion engine if you prefer your models to have a physics slant). If we can maintain a small ‘improvement gradient’ over a longer timescale we can keep all teachers developing in a sustainable way. On any given day the improvements feel small, but over a year they aggregate into permanent positive change. Desirable difficulties in action.

Creating a culture of continuous improvement

Creating a culture of continuous improvement requires all staff to participate. Often our best are keen to get better and our worst need to improve, but our average teachers get ignored as they drift along.

If we can establish a culture within our department that makes all staff active participants in their own improvement, it reduces the possible negative consequences of a more formal ‘stages of support’ for weaker teachers. From their perspective they are just being supported like everyone else. To the Head of department they are top priority, they get more frequent learning walks, more frank conversations, less choice of next steps. The improvement needs to come, but it will come without the loss of morale and will deliver a more permanent change.

Sustainable improvement through incremental coaching is often the aim of a lot of school learning walk policies. But actually getting it to work is incredibly difficult. It takes time and doesn’t fit easily within a 6-week time frame.

How do we build capacity within department leadership?

For this to work it has to be a priority. It needs time. I honestly think the choices a head of department has to make when it comes to prioritising their time are incredibly difficult. But consider this: if you do nothing else but increase your teaching staff’s teaching ability by 10% then you have made a huge impact on all the students in your school.

To avoid the very real drag towards admin and crisis management I have found it best to diary in blocks for learning walks. You will see a few ways outlined in the plan below.

How do we build capacity by using the Senior Leadership Team?

My SLT link is also expected to do learning walks to quality assure the teaching in science. Unfortunately while he is exceptional in many ways, he is not a scientist. This means asking him to give advice on the subject-specific aspects of teaching is relatively pointless. But fortunately, teaching has a large amount of generic aspects which he can help with.

Routines and generic classroom practice are areas that all leaders can support on. Most teachers have a certain aspect of routines or classroom practice that can be improved. As long as the school has a shared language of common classroom features (personally, I like TLAC and ‘get better faster‘) the senior leaders can support staff in developing effective classroom habits within the whole school’s ethos.

How I intend to create this culture:

  1. I am going to devote 2 hours a week of my leadership time to learning walks.
  2. I am going to direct my fellow department leaders to devote 1 hour a fortnight to learning walks.
  3. Every meeting I have with a leader will be no more than 30 mins, the second half will be a paired learning walk.
  4. I will ask my SLT line manager to do drop-in visits to staff to check they are using TLAC and our school behaviour system correctly, including identifying any reluctant learners.
  5. Every learning walk will have a face-to-face discussion with the member of staff and one target only identified. If they are part-time then mutually consented email might be required.
  6. We will keep a shared log so I can see at a glance what issues are arising. If there are global issues, or problems due to systems, these will fold into our department time.
  7. Staff will be visited by the same leader for a half term. This aims to create a collegiate approach and make it easier to practise one specific target over a period of time.
  8. I will operate an open-door policy where I will welcome staff to watch me teach at any time. One team leader will be directed to learning walk me on top of this. No one is above the group.

Let me know your thoughts; Do you think I’ve missed a trick? Do you have any questions?


I have not discussed how to observe, what to look for or how to give feedback. They would be large blogs in themselves. Maybe in the future.

*I have been that teacher. I have had times in my career where I have been considered an excellent teacher and other times when I was considered weak. In both those times I don’t think I was that different in my teaching, but the criteria I was judged against was massively different. The things that did improve my teaching: CogSciSci, ResearchED and twitter.

**I know, it’s a bad joke and too obvious but consider this: It’s my blog. Also I give you this amazing video of Bjork explaining how a TV works as an apology.

***Don’t come at me quoting Fick’s law! I know the concentration gradient is important in diffusion. But its not as important as the ability to maintain it. This is why evolution produced awesome counter current systems to ensure that even in times of plenty there is optimal diffusion occurring.

Comments are closed.

Blog at

Up ↑

%d bloggers like this: