I’ve seen a fair few school improvement plans and have never really been completely satisfied with what I have produced. Each one is iteratively better than the last, I hope, not in how they look but the thinking that sits behind them. However we think of them and however they look though, they do represent a common problem in education and school leadership of treating the complex as complicated. We’re trying to improve provision in our school so that children leave as well rounded individuals and there are multiple influences on and barriers to this. When you throw in regulator check ups such as Ofsted inspections, the need to try and prove cause and effect or to create improvements in aspects of an inspection framework also warp our efforts. The narrative that school improvement plans often reinforce is one of linear cause and effect; we give ourselves the illusion of control about how we make a difference to the school.
Common features of school improvement plans are a goal, actions to carry out, criteria by which we measure success and some sort of accountability nod (usually a name next to an action). The boxiness of these plans can be hindrance to the actual work required to improve our schools, with plans easily becoming underused, out of date or redundant.
What I’m proposing are a couple of tweaks to shift how we think about what we’re doing. Do these tweaks reduce the boxiness of school improvement plans? Certainly not but we’re never going to capture the full complexity of school leadership in a table. The best we can do is make sure that it guides thinking towards appreciation of complexity rather than working completely in opposition. The tweaks are:
- A goal hierarchy
- A range of success criteria
- A process for checking leaders’ understanding
- A process for filtering actions into middle leadership
A goal hierarchy
It’s not easy to choose suitable goals. We have all sorts of information at our fingertips that we could use to choose an improvement priority: attainment and progress data, inspection recommendations, survey data, and national or governmental agendas. But each of them provides only a limited insight into the full reality of the school. Ask each colleague, child, governor or inspector what the school needs to do to improve and you’ll get a fairly long list, each influenced by their day to day experience.
And when we settle on a goal the level of specificity needs careful thought. Too broad and we struggle to know where to start. Too narrow and and we risk not capitalising on complex interactions.
So, a goal hierarchy:
In this example, there are so many possible elements of meeting the needs of children with SEND but we’re choosing a smaller number of them deliberately, based on the various data that we have. The broad improvement area is general and should be an easily understood anchoring concept to develop meaningful learning among those enacting this plan.
Why is a goal hierarchy important? For this I refer to meaningful learning. A school improvement plan is not just a blueprint for how we’ll improve our school but also a tool for teaching the team how to think. The cognitive load of thinking about school improvement, let alone communicating it in a way that renders is comprehensible to a wide range of colleagues (each with varying prior knowledge and underlying beliefs) is significant. I think it is useful to track complex concepts back to their most fundamental, overarching of concepts because the more we do that, the more likely it is that we find common ground on prior knowledge and therefore a solid base on which to build collective understanding and in turn coherent action.
A range of success criteria
We’re often held to account with numerical or categorical values. Attainment, progress, behaviour and attendance data; survey data; inspection gradings. We risk treating the complex as complicated if we set ourselves up to believe that there are simple cause and effect relationships between the things we do and the effects we see. Those numerical and categorical values are often quite distal from the attempt to improve them, coming to fruition months or even years down the line. Metrics can be tyrannical and have a significant influence on trust between colleagues as well as encouraging gaming. While we need to have an eye on them, it is probably more worthwhile describing qualitatively the future ideal of how we want the school to look. This serves two purposes. The first is bringing narrative into our school improvement endeavours. Stories are psychological privileged in that they are easily understood and are memorable which makes the sharing of why we’re doing what we’re doing much easier. The second is we can check our progress of how we’re doing more easily than relying on distal numbers that average out important nuance. We can seek to compare the reality with what we wish it to be.
Example. Remember our goal hierarchy?
Well, what does a future ideal look like regarding meeting the needs of children with SEND?
The assess, plan, do, review cycle is scheduled multiple times across the year. The SENDCo keeps teachers informed of the stage that each child is at, organising timely meetings to discuss children’s progress. Teachers use a range of data to assess progress towards targets. Teachers come prepared to meetings with examples of how children have met targets and with an idea of how targets need to evolve. Teachers keep parents well informed of their child’s progress.
That’s just a few thoughts on the first element. Not much of this description of a future ideal could be adequately captured with metrics, although some metrics could be useful as an indicator of how well we’re meeting needs. For example, the number of children on the SEND register might increase because as a result of the work we’re doing, we’re better able to pick up children with additional needs. It might also decrease because we’re doing it so well that children no longer need support plans.
A word here on the imagined school. If you ask most headteachers to describe how the assess, plan, do review process works in their school, they’ll probably describe the above in some way. But do we know for sure? We can’t. We rely on the intent that we have communicated, the feedback from colleagues and parents, as well as the odd sampling of how it is going. Setting out a description of a future ideal gives us a better focus for testing the reality of school life than metrics do.
A process for checking leaders’ understanding
Sometimes we don’t talk enough about the thing that we want to achieve and do and as a result, leaders in the same team can have quite different understanding of them.
A caricature of this problem might be the headteacher writing some actions about what should happen to address school improvement priorities and that these are delegated to another leader. They read or listen to the detail around it and, inadvertently, have a different understanding to what was intended. Time could then be wasted on poor implementation and if the issues persists further as it is filtered down the school, there could be huge gaps in understanding. If we’re committing to an action, sufficient time should be spent exploring understanding of it. So when we’re crafting actions, I suggest a way of checking understanding that includes a RAG on the actor’s part of the extent to which they need to discuss an action. This can be a good way to focus 1:1s and to keep a school improvement plan alive in conversation from week to week.
In this example, the SENDCo’s 1:1 with their line manager can be focused on the support that they need to enact the plan. But remember we’re notoriously bad at judging our own competence and these would all need exploring to a certain extent.
A process for filtering actions into middle leadership
School improvement plans are at risk of never making it out of the SLT. We might have a school improvement plan, an English plan, a maths plan, a SEND plan, a disadvantaged plan, a safeguarding plan. The list goes on. And while different areas of school life might need their own improvement plan, the overarching school improvement plan needs to bring coherence or we can have conflicting and contradictory actions, not least on colleagues’ time and attention and teacher burnout because everything feels like a priority.
So let’s have a process for filtering down. For each action, where does it go? Does it stop here as an action for senior leaders only, where no-one else needs to do anything? In the SEND example above, action 1 kind of stops, but the other two could benefit from more connection with wider school work and aligned effort from other leaders.
Back to our SEND example.
For action 2, it is all very well modeling this process for teachers but there will definitely be some subject knowledge gaps somewhere, with some teachers, that stops them from doing the plan part well. So the SENDCo leading this action might need to work with the Heads of Subject (particularly English and maths) to support this work. Maybe this action is filtered down into the English and maths action plans to collate some example small step progression for common areas of need. This links in leaders beyond the SENDCo to create coherence in what leaders across the school are focusing on. This isn’t a passing on of responsibility, but a reminder and encouragement to make connections; improvements could be stickier as a result of a coherent approach. Similarly for action 3, Heads of Subject could be instrumental in exemplifying a range of data to assess progress towards targets with their deep subject knowledge and again, this could be built into their action plans for coherence.
The example above also references Head of Year and Head of Department timelines but that’s for another post.
Here’s a suggested template in it’s entirety:
Is it still boxy? Yes. But might it help to teach those that use it to appreciate complexity and reject the linear cause and effect narrative that doesn’t;t accurately represent how schools work? I hope so.
MrNickHart 