The below was sent to me by a friend who wishes to remain anonymous for fear of ramifications if their senior leaders see it. Please read, enjoy, and show them some support. 


Humans are intentional beings. We want things to happen and we choose actions to bring those things about.  But there is folly in judging the goodness of something by the intentions of its agent, rather than its actual outcomes. “For-profit” companies become insolvent. Political regimes are established to liberate but grow to oppress. The violent alcoholic partner who really wants to change is still a danger to you and your children.

The aim of differentiation is to help children. Because children enter our classes with different starting points, it is rational to think that by giving them different work, we can challenge each group the correct amount and allow each child to make the most progress possible. The intention is good.

But what is the outcome? The outcome is that for the past 15 years, observers of lessons, including Ofsted, praised our lessons for catering to the needs of all groups, and we go home happy knowing that we have been good teachers.

Stop! That is not a proper outcome. What is the effect on learning?

I have tried and failed to find any research evidence that links differentiation to increased outcomes for students. This blog by Greg Ashman is a thorough discussion of the evidence that does exist, and shows that none of the research conducted shows that differentiation improves learning.

This is quite horrifying. The opportunity costs for time spent planning for differentiation are huge. It begins to look as though differentiation could have the opposite outcome to its intention, like those bankrupt companies, totalitarian governments, and abusive husbands.

In fact there is evidence to show that this is exactly what happens. Greg Ashman again, this time plotting mean test scores against amount of differentiation from TALIS data:

ros 1

But how can this be? We are planning to help children, not hinder them: how can our plan not succeed.

The first thing to say here is that just because you want something to happen, doesn’t mean it will happen. This sounds ridiculously obvious but I don’t think many people think about it in education. Things happen because of mechanisms, variables, the physical nuts and bolts of the initial conditions and operations.

Let us consider the nuts and bolts of differentiation. In one form of differentiation, teachers create for example three activities of varying difficulty and students choose which one they want to do. This is pretty obviously going to inhibit progress as students with aspirational backgrounds choose challenging work and those without choose easy ones. The gap widens.

A perceived improvement on this is for the teacher to look at previous assessment data and then assign the different tasks accordingly. But this approach represents a serious misunderstanding of data. The data points we have for individual children, from KS2 tests, internal assessments and the like, can only tell us about the particular questions on that assessment they did did well on and did badly on. You can’t put a ceiling on a student’s work on Romeo and Juliet because they did mediocre on Of Mice And Men.

Ok, how about we put students into groups based on their question-level analysis or performance in front us on mini-whiteboards, and they work on the areas they need to improve? This is deceptive. It sounds like it solves the problems in the first two examples, and in theory it does. But in practice we have a different story.

Education is a practical activity. You have real students, 30 of them, 30 chairs, a room, a whiteboard, a teacher. Putting students into groups and getting them to work on something independently is never going to give you the outcomes you would get from interactive teaching of the whole class, directed by the teacher. It is never going to happen. Students need guidance from their teacher. That is a fact of education. You can run around the room giving rushed guidance to each group. Or you can give well-delivered, confident, deliberate instruction to everyone. The objection here is that students from group 5 will have to listen to your explanation of question 2 even though they all got it right. This won’t be an effective use of their time, is the argument.  Better put everyone into groups again and get them going on their own, right? Wrong.

The difference here is intention vs outcomes:

Whole-class teaching:

ros 2

For differentiated groups, students have to listen to the instructions about the groupings and the activities, worry about what their grouping means and whether they have got friends in their group, move into their group, retain the instructions in their head, or work them out again if they are written down… and then they have to do the work, either on something that is new to them or the thing they struggled the most with and got wrong before, with minimal guidance from the teacher. Even in the most perfectly organised, impeccably behaved classroom, in this set-up, the teacher must divide their attention between groups. If there are 6 groups, in an hour’s lesson, each group will get a maximum of 10 minutes of teacher attention each. And this is not 10 minutes of precisely timed support, just when the students need it. It’s 10 minutes when it can fit around the other 5 groups. This is crazy. No student will make the most progress possible under these conditions. And the students who need the most support will struggle the most and fall further behind. You can pretend to yourself that this is not true, that if differentiation is done really, really well, this is not the case. You can lie to yourself if that is the path you choose. But outcomes in the real world happen because of real things, not just because you want them to.

We should support our learners by creating excellent explanations and practice questions, by peppering them with questions and making sure they work hard all the time. We should give all students the best possible opportunity for success by teaching them in a unified group, with rational acceptance that 5 minutes might be revision  or easy practice (which in any case are valuable for the building of long-term memory) so that 55 minutes can be effective learning for all. We should not indulge our vane delusion that we can make something happen just by deciding that’s what we are doing, and never mind the laws of time, space and cognition.

Ofsted is no longer proscribing preferred teaching methods. The focus for quality of education, apart from external exam results, will be the quality of the curriculum, and its intent and implementation. We can expect Ofsted to be asking a lot of questions about how curriculum is implemented. My main question here is: “Is a well-planned differentiated lesson as effective at implementing your curriculum as a well-planned whole-class teaching lesson?” The answer is no and it will always be no. You can’t change space and time. You can’t make things happen just by wanting them.

And schools up and down the country are getting brilliant results with whole-class teaching, much better than we’ve ever seen come out of a differentiated lesson. The work by students at MCS Brent and Magna Academy Poole is phenomenal and is the result of carefully planned whole-class teaching. And if you’re into Ofsted reports, they are both Outstanding.

Some people will read this post and then pretend to themselves that they haven’t read it, or rationalise it in some way so that they don’t have to change their outlook. It’s uncomfortable to realise we got it wrong. The realisation will come, in maybe a year, maybe two or three, when Ofsted come and their question is “Is this the most effective way of implementing your curriculum?” Many will not want to wait until that point, and will face this truth today. Differentiation was a mistake, it sounded great and we meant well but there are fundamental reasons why it always fails in comparison to whole-class teaching. We are teachers: we are here for our students and our subjects and we’re prepared to change our minds if it means better outcomes for all.

Once you have read about these arguments, you can never go back. You can pretend to yourself that you haven’t read them, or you can rationalise this situation to yourself, protect yourself from the truth, but you can’t claim integrity. Or you can turn your face to the sun and march against the myth. It takes courage and humility to admit that we got this wrong but it is the only morally and intellectually sound thing to do. Not everyone has got what it takes to do this. Have you?


Adam’s note: I think it is a crying shame that some leaders are so small minded that their staff don’t feel comfortable publishing articles that challenge the orthodoxy. I have found writing to be an incredibly helpful experience for myself, and because I don’t think that anyone should be denied that opportunity for arbitrary reasons if you want to write an article to be published here anonymously feel free to send it to me and I will do my best to help you out.