Making School Development Work

School development planning could be so much better if more people knew about Problem Driven Iterative Action…

At the start of this half term, I read this excellent blog by Harry Fletcher-Wood on an approach to improvement planning called Problem Driven Iterative Action. The blog was set reading as part of the Steplab Certificate in Coaching Leadership (a superb course that anyone interested in teacher development should try and take!) but since I read it, I have been thinking a lot about the logic and sense that seem to underpin it, so much so that I have asked the rest of SLT to read it in advance of a discussion about how we write our School Development Plan for the next academic year. While reflecting, I also realised that we’ve actually inadvertently applied PDIA to a change to our Writing curriculum during this academic year and I think it might explain the success we’ve had with this particular change and so, it has become the subject of this blog. First, what’s the problem with big plans?

What is the problem with big plans?

Big plans feel good. In his blog, Harry Fletcher-Wood wrote that ‘when we face a problem, our instinct is to come up with a grand plan to solve it’. They make the future seem predictable, they make us feel like we are in control and that we are strategic (which is one of those words that we seem to be increasingly evaluating the quality of leadership against without a shared understanding of what we mean by it!) The problem with big plans that have lots of detail and lots of moving parts is that they rarely behave as we expect them too. Perhaps part of the reason for this is that schools are complex places and so school improvement is necessarily a complex system. Let’s pause here to consider what we mean by ‘complex’. Complexity theory is another fascinating world and one I hope to dive into in greater detail over the course of the summer. The brilliant Dr. Neil Gilbride is one of the leading thinkers in this area and I know that he is doing some really inspiring work with school leaders around the ideas of complexity. From my initial reading of a small part of the complexity literature, one feature of complex systems really resonates – emergence. This is the idea that within a system, you have various elements that interact to create an outcome. However, the outcome itself is greater than the sum of its parts and once it has ‘emerged’ it continues to develop and evolve separate from the continued interaction of the parts that created it in the first place. I think this is what happens with school improvement initiatives. We carefully plan out how the different elements will interact and combine to create the outcome we want but along the way, many unpredictable and unintended interactions also take place and give rise to unexpected evolutions. 

What is the solution to this?

The problem with school improvement is that we can’t possibly predict all of the challenges that any given initiative will throw up. The bigger problem is that so often, when a problem does appear, the initiative is already operating at scale and involving so many people that solving that problem becomes really difficult. For example, imagine a new behavior policy that a school role out in September. The roll out is accompanied by a set of actions assigned to different post holders. SLT will patrol the corridors, class teachers will be responsible for using the new policy exclusively, middle leaders will be running weekly behavior reports in their phase or subject to analyse the impact of the change and the behavior lead will be looking at whole school level data. That’s a lot of people with a lot of operational tasks and time invested into this new policy. That has two problems:

1. Energy. People who invest their energy into a new system may start to see their own part of the system as the thing by which to judge its success. For example, ‘I run my reports every week and it helps me see which classes I need to put support into’.

2. Time. With everyone having to invest time into making the new system work, their time to think is diminished. This means that it becomes harder to see the problems that exist. 

   
   

Investing heavily in time and energy at the outset of a mass roll out also makes it harder for leaders to change course if they do recognise that problems have cropped up that they weren’t expecting. It can also encourage leaders to normalise and ignore the challenges that are faced. For example, ‘Oh there were always going to be teething problems, we just need to stick with it’. Before we know it, we are 6 months down the line and it feels that we have invested too much to possibly peel back and solve the problems at the very route of the initiative and so we plough on, often with unspoken discontent that gradually erodes the remaining energy that people have for making the change work. How can PDIA offer a different way?

What is PDIA?

Taking a PDIA approach looks and feels different for everyone involved. Leaders and staff work together to identify a problem. They then identify a small group of people who will trial an initial solution, not just to see if it works, but rather to find out where the problems are at that really localised level. This is where the iterative part comes in. The small group come back together and discuss what they have learned from the initial trial. With that learning in hand, they run the initiative again and then come back together again. This repeats until they are satisfied that they have found an iteration of the plan that works at a local level. 

In the case of school improvement, I think that before a plan reaches whole school roll out, it should go through a kind of middle ground of testing. This involves the plan scaling up and involving more groups. This time, when they come back together, there are less localized issues (they were identified and addressed in that first phase) but there are a different set of issues, the kind that we find when multiple phases and departments work on the same initiative and the initiative necessarily adapts (another principle of complexity theory) based on the characteristics of the department or phase. For example, that same behavior policy that works for Year 5 and 6 will need significant adaptation to be fit for purpose in Reception. Staff teams need the time and space to discuss these adaptations together and then bring them back to the wider group to agree on them. This is important because certain aspects of the initiative need to be close to the initial design and others aspects can be more readily adapted. This is what has become known as the ‘active ingredients’ of an initiative. 

Once the initiative has been iterated and adapted at that mid-level, it is likely that leaders will have a version of the it that can be rolled out across the school with the anticipation of the most likely wider issues but crucially, with a team of people who have already experienced and become invested in the change to help drive it. With that, the chances of success at whole school level are greatly increased. 

What did we do? 

Back in September, we knew that something wasn’t quite right with how writing was being taught. Student achievement in writing across the school lacked behind reading and maths and when we spoke to students, they were almost always less assured in their responses than in other subject area. But, we didn’t know exactly what the problem was. So, together with the writing lead and the reading lead (who is an experienced KS1 writing teacher), I sat down and we thrashed out what we thought the problem was based on the information we had available to us. We settled on the problem being that students didn’t seem able to write independently with stamina and an assurance behind their writerly choices. We also decided that the problem could be traced back to a lack of live modelling and explicit narration of thought processes, choices and rejections from teachers. 

Once we knew what the problem was, we moved to designing a small-scale solution. This involved the writing lead trialing a new approach to modelled writing that we had designed specifically with the goal of better preparing students for independent writing. The approach really sought to highlight the importance of intent and purpose in writerly choices so that students would start to see that the choice they had as writers was not to be measured in a binary right or wrong manner but by degrees of effectiveness. The writing lead then taught the next unit of work to their class and implemented the agreed approach. After two weeks, we came back together to discuss the problems that had emerged. 

The first problem that was reported was that students were not applying the purpose and intent concern to longer pieces of writing at the end of the unit. To address this, we tweaked the approach and added in a number of what we called ‘short burst’ writing sessions in the unit of work. These were designed to give students experience at writing for a purpose over a mid-length piece, helping them bridge the gap to independent writing. So, off went the writing lead again and we came back together in another two weeks with much more positive feedback about the effectiveness of what he had done. 

Our next step was to test out the approach in Key stage 1. So, the writing lead and the reading lead (the experienced KS1 teacher) worked together to plan a unit of work for her Year 1 class. When we came back together two weeks later, we found that the idea of purpose and intent wasn’t landing anywhere near as well with the younger children. We considered that this was likely to be because it was taking so many more cognitive resources for them to formulate letters, words and sentences that there was little left to think about the choices they had. So, we tweaked our plan again for KS1 and had the focus of the modelling be far more around subject and verb agreement and modelling sentences as one complete thought. When we came back together again two weeks later, we found again, more positive results. 

At this point, we rolled the approach out to the whole of Year 6 and Year 2 (6 classes in total) and later to Year 4 (another 3 classes). Each time we scaled up the approach we found new problems but we also found that our collective mindset was becoming increasingly positive about solving those problems. After six months of iterating, we finally got to the point of rolling out the new approach across the school. 

What difference did PDIA make? 

While it is impossible to know what would have happened had we rolled out this change on a whole school level from day one, what we do know is that it felt far more strategic and collaborative as a process. Enabling small group or individuals to really drill down into the inevitable issues with the approach meant that we didn’t have to worry about the wider impact of the same thing going wrong on a larger scale. We were able to spot the problems and solve them quickly. This meant that when we got to whole team roll out, we had a really refined approach and crucially, 9 teachers out of 21 who had already implemented the new approach to some extent. This made transitioning to it being a whole school approach much easier as we not only had a strong force of positivity but we also had almost enough teachers for a 1:1 mentoring scheme to implement the new change. 

Closing thought 

Change is always difficult in schools. Improvement is even harder, not least because of the complex nature of a school system. PDIA embraces that complexity and works with it rather than trying to circumnavigate it with a grand plan. Our development plan for writing didn’t look fuller than it did the previous year, we didn’t know on day one what actions would be assigned to who… but through the series of problem driven iterations, we adapted our approach until we had something that worked for all of our students. The process we followed meant that we had to wait before we could see whole school impact but sometimes, we have to trust the process rather than try to prove the process.