Good differentiation in science is hard, perfect differentiation is impossible and quite a lot is probably ineffective and perhaps even damaging to learning. Let me try to explain why.

#### What does good differentiation look like?

Let’s start by thinking about what good differentiation looks like. Differentiation is about making sure that all students are appropriately challenged. As John Hattie suggests “if you are not challenged, you do not make mistakes. If you do not make mistakes, feedback is useless.” Too hard and students will not get the intrinsic satisfaction of solving a problem. Too easy, and students will get bored and fail to develop. Differentiation is not about lowering expectations of what can be achieved, we just need to vary the approach of how to get there. So how can we appropriately challenge all students in our classroom?

#### What is not great differentiation?

I don’t think the answer is to give every student an individual worksheet. Blue sheets for students graded C/D and pink sheets for students graded A/B. The reason this doesn’t work is it assumes the teacher has an accurate and complete understanding of where each student is *on that topic,* before the lesson begins. Equally asking students to chose the appropriate worksheet is difficult as it relies on them having an accurate idea of their own ability and this is often wrong.

Defining objectives for all, most and some is admitting defeat from the outset. Instead let’s raise our expectations and improve our teaching – what is good for ‘low performing students’ is likely to be good for all.

#### Let’s learn together

Another, perhaps more important problem with individualised approaches to differentiation is that we lose the opportunity to learn together; for me, this is the point of school. Children learn when they learn together (Vygotsky) and the more we break up tasks the more we fragment and temper this hugely beneficial way to learn.

#### How can we do differentiation better?

What’s the solution then? I think the best differentiation is twofold. Firstly, it requires teachers to adapt the teaching in the lesson in real time for individual students – this needs good checks for understanding. If students are finding the work easy, then the scaffolding is removed. If students are struggling then get modelling and adapt your teaching approach, not your expectations of what can be achieved – differentiation is not about lowering the bar. Secondly, make sure you have a proper understanding of progression for the concept you are teaching; this will allow you to select which of the knowledge and skills that *could* be taught *should* be taught.

#### Some practical tips

- When you create tasks make sure they build in difficulty and are not too short. If tasks are short, then it means all students must move at the same rate and this doesn’t give you opportunity to help those who need 1-1 support
- Use the white board to teach responsively. For example, if some students are struggling to balance equations, you could set most the students working on a worksheet and then support the rest using the white board
- Be mindful of falling into the trap of thinking that very able students don’t need to learn the basics first or that low attaining students don’t need to apply/extend their learning.
- If you create the right culture, you can reveal answers as students complete questions. This allows students to know they are on the right track and so can move at a more accelerated pace.

##### Further listening and reading

- John Hattie onThe Educators, BBC Radio 4 (2014). What really works in schools and classrooms
- McLeod, S. A. (2014). Lev Vygotsky. Retrieved from www.simplypsychology.org/vygotsky.html
- https://educationendowmentfoundation.org.uk/resources/teaching-learning-toolkit/individualised-instruction/

- Planning lessons: the EPIBA approach
- Clearly defined lesson objectives
- The Do Now
- Activate prior knowledge
- Challenge your students
- Use a context
*Challenge all students appropriately*- Use direct instruction to provide clear explanations
- Model abstract ideas in concrete ways
- Use questioning to probe understanding
- Check for understanding – give and get feedback
- Troubleshooting – why did it go wrong?