Differentiation? I hate the word as I hate Hell, all ludicrous bureaucracy, and thee!

Er… Shakespeare, Romeo and Juliet

Differentiation is one of the darkest arts in teaching. The generally accepted position is that differentiation is wholly good, and this is the cause of the wracking guilt felt by harrowed teachers: it may well be good, but it’s bloody hard work. My bottom line is this: any policy predicated on the idea that teachers should work harder is doomed to failure.

Thankfully, teaching’s enforcement arm seem, at long last, to agree:

“It is unrealistic … for inspectors to necessarily expect that all work in all lessons is always matched to the specific needs of each individual.”

Subsidiary guidance supporting the inspection of maintained schools and academies, December 2013

Tom Sherrington writes in his hugely popular post on differentiation that, “every class is a mixed ability class so, regardless of our views on selection or setting, all teachers need to cater for students with a range of skills, aptitudes and dispositions.” But does this mean we have to flay ourselves producing individual lesson plans for all the uniquely different little blighter we teach? Like many of the slippery terms used in education, differentiation can mean all sorts of things depending on who’s talking and in what context. Does it mean coping with difference? Learning for all? Success for all? As the landscape’s changed over the past few years, there’s an increasing consensus that ‘success’ should be differentiated: our examination demands winners and losers. Where does this shifting terrain leave us? Claiming that differentiation just means we’re special and different and should be treated as such is bland to the point of meaninglessness.

So, let me offer my own definition of differentiation: Getting all pupils to do something they find really hard. Increasingly, I see our job as being not to make work easy, but to make it as hard as possible so that pupils will make mistakes. Mistakes are the very stuff of learning. If your pupils aren’t making mistakes and struggling, you’re not doing your job.

Show me a teacher who doesn’t fail every day and I’ll show you a teacher with low expectations for his or her students.

Dylan Wiliam

Obviously, we’d all want to argue that we have high expectations for our students, so it’s helpful to think about what we do in this way. Teaching to improve short term performance is teaching that seeks to make things easy for our pupils. Teaching that explicitly seeks to improve long term retention and transfer of skills and knowledge is teaching that will result in pupils making mistakes.

Consider this example: The chemical symbol for lead is Pb. What’s the chemical symbol for lead? Of course it’s a caricature, but like all caricatures there’s enough truth in it for us to recognise something naggingly familiar. This is how we’re encouraged to teach, and it’s designed to minimise mistakes. There’s no real progress in knowing the answer to a question you’ve just been told. This then is an example of the low expectations we should aim to avoid.

Teach to the top, support at the bottom

Pupils should be expected to get over the same bar, but will need different ladders*. Effective differentiation aims to start with end point and plan how to get all pupils there. This depends on three things: 1) the quality of routines and relationships, 2) the teacher’s commitment to explicit modelling and scaffolding, and 3) marking.

1) Routines and Relationships

The better your students know what to expect and the better you know your students, the better your ability to differentiate will be. Routines need embedding. They take time and effort to embed. Spending time teaching pupils how to enter the room, present work, respond to questions etc. may seem trivial, but they are groundwork for everything else. When routines are established, relationships can grow. But just knowing their preferences and idiosyncrasies is not enough; you should also know the data. I’ve blogged before on how data can be made meaningful. This much is, I think, obvious.

2) Explicit modelling and scaffolding

One sure-fire way to demonstrate low expectations is to rely on success criteria: they are nothing more than a terrible checklist of low expectations. If we want our pupils to produce high quality work then we will need to provide them with exemplars to deconstruct and commit time to modelling the meta-cognitive processes an expert engages in. Year after year, I’ve watched Wimbledon without getting any better at tennis. How can this be? Unfortunately, we don’t learn well from watching experts work. I only started to get better at tennis after taking lessons and having the processes broken down so that I could recognise and understand what I should be doing. If we fail to model exacting standards, pupils will fail to achieve them.

Scaffolding is the art of knowing what someone is capable of and then supporting them to do something beyond their current capabilities. The trick is remove the scaffolding as quickly as possible so that pupils don’t become reliant on it. This is the problem with writing frames; the scaffolding is hard to remove. You can go through the process of using Black Space to increase lexical density, but it’s far simple and much more straightforward to provide the scaffolding at the point of speech. For reasons that are mysterious to me, it’s a magical fact that pupils are able to write what they can say. If we prompt them to ‘speak like an essay’ they’ll be able to write like an essay. “Oracy,” as James Britton said, “is the sea upon which all else floats.” If you can say it, you can write it. You can read about this in more detail here.

3) Marking is differentiation

I’ve made a bit of a name for myself by banging the ‘marking is planning’ drum, but I’m also convinced that marking is the most effective way to differentiate. Seeing what mistakes our pupils have made and then giving them specific feedback which they are directed to act on in lesson time is the only sane way to ensure that pupils do have work with is matched to their specific needs, regardless of whether it’s necessarily expected Ofsted inspectors.

And proofreading is an important part of this process. If you’ve embedded the routine that if work isn’t proofread it isn’t marked, then pupils will become skilled at spotting their mistakes. The mistakes we spot ourselves are the ones we’re most likely to learn from. It’s all very well the teacher pointing out where you’ve gone wrong and providing instructive feedback, but it’s so much easier and more beneficial if you can do it yourself. The practice of engaging meta-cognitively with your own work is hugely powerful lever for learning.

And that is how I think we should best approach differentiation. Not as a back-breaking exercise in producing teetering piles of pointless paperwork, but by having consistently high expectations of every student we teach, regardless of their ability and by encouraging them to make, and learn from, their mistakes.

*@TWDLearning came up with this neat summary.

Related posts

Differentiation: to do or not to do
Building challenge: Differentiation that’s quick and works
Differentiating the responsive way from Andy Tharby is also good.
And, surfing the differentiation zeitgeist, Tom Sherrington synchronously posted this today.