The tyranny of the progression statement

The assessment approach that never dies

Regular readers will know that we are not keen on progression statements!

These are the collections of sentences like “I can infer characters’ emotions from explicit details in the text” which chart the progress of students through a curriculum.

The problem with these statements is that they are too vague and imprecise to provide useful information for a teacher or student, as you can see below.

In England, they were a feature of the first national curriculum in the early 1990s. Almost as soon as they were introduced, they were criticised. One of my favourite critiques is from Christine Counsell. She would take a set of ten progression statements in history that were meant to run from year 1 to year 11. Then she would ask people to put them back together in the right order. No-one could do it accurately, because they were just word salad.  

Then in 2013 national curriculum levels were finally abolished, and we all thought we could move on.

Except – over a decade on, even though there is no statutory requirement to use this form of assessment, and there are a whole cohort of teachers trained with no memory of national curriculum levels, they are still really popular.

A couple of weeks ago, the primary teacher and reading specialist Christopher Such said that the most common question he gets asked is:

• “What should our school put on its reading progression documents?”

People want sets of statements like these.

• Year 4 – Inference: be able to infer characters’ emotions from explicit details in the text.

• Year 5 – Inference: be able to infer characters’ emotions from implicit references in the text.

Christopher goes on to explain all the problems with this approach from the perspective of teaching primary reading.

If a pupil in year 2 recognises that Winnie the Pooh is confused to find Piglet in his house, is this pupil working at a year 5 level for inferences of this sort? Equally, if a pupil in year 7 struggles to recognise Mr Darcy’s repressed ardour towards Elizabeth, does that mean that they are working below this level?

So what is going on here? Why has such a flawed and unloved assessment system managed to stagger on for decades even in the absence of any statutory requirement to do so?

Not only that, but this approach is also common in Scotland, Wales, the US, Canada, Australia and New Zealand – all countries with very different education systems to England and to each other.

Why?

I think these systems become popular because they are a logical outcome of thinking that skills can be taught directly. If you think that inference is a discrete skill, then a statement like “can infer insightfully” makes sense. Once you realise that inference is not a discrete skill, and that your ability to infer depends on vocabulary, background knowledge and text complexity, then you realise why these systems don’t work.

I think these systems stay popular even once their flaws are revealed because of the lack of good alternatives. It’s easy to say that statements don’t work, so just ditch them, teach a good curriculum and don’t bother replacing them with anything. But I have never felt this is an entirely satisfactory answer. I think schools and teachers are right to want some way of measuring their students’ progress through a curriculum.

But it is not easy to change to a new, better system precisely because these statements go beyond mere assessment and have huge implications for the curriculum & pedagogy.  If we accept that one of the main problems with statements is their vagueness, then in order to improve them we need something more specific. And as soon as you get more specific, you then need to be more specific about what is taught on the curriculum. I think that’s a good thing, but it’s clearly not an easy thing. And whilst a spreadsheet with a list of statements on it is easy to share with colleagues in another school, a specific curriculum is much harder to share and adopt.

If you do have a curriculum that you are happy with, then there is a relatively straightforward way of making statements more precise and more useful, and that’s to back them up with sets of closed questions.

So you would take the statement “Can compare two fractions to identify which is larger” and back it up with ten questions. Setting these questions would force you to think about exactly what you mean by this statement at this particular point in the curriculum. Do you mean “which is bigger: 3/7 or 5/7?” Or do you mean “which is bigger: 5/7 or 5/9?”

In other cases, backing up the statement with questions will be harder. What questions would you design to test the statement “able to infer characters’ emotions from explicit details in the text”? In order to test this statement, you’d have to ground it in an actual text. You’d have to decide whether you mean Winnie the Pooh or Pride and Prejudice. And that’s a good thing. You might also realise that the ability to make such inferences is dependent on knowledge of vocabulary, and so it might be worth teaching and assessing vocabulary explicitly too.

If you have a good curriculum, this process should be relatively straightforward. If you don’t have a good curriculum, this process will probably expose that. As Dylan Wiliam says, “assessment operationalises curriculum”. Flawed thinking about the curriculum is often only exposed when you start to design the accompanying assessments and analyse the results.

Backing up statements with questions grounds them in specificity and reality and creates a genuinely common language – not just the illusion of one.

At No More Marking, we have created a Writing Progression sequence where every statement is attached to a short quiz. You can view it here. The full sequence is available for subscribers, but we have made the first 3 statements available for everyone. You can also watch a recording of a Writing Progression information webinar by going to the “Previous” tab at this link.