It’s been 27 years since Shirley Clarke began her work on formative assessment, and she’s as passionate about its place in schools as ever.

But she’s frustrated, too. Since being popularised by Paul Black and Dylan Wiliam in the 1990s, the term “formative assessment” (which was originally coined by Michael Scriven in the 1960s) has come to be used so widely that many people in education “think they know it when they don’t”.

This has led to schools cherry-picking parts of the formative assessment framework Clarke developed, and expecting it to transform outcomes. 

“It’s no good saying: well, we’ll keep these things the same, but change others. We’re still doing formative assessment. No, you’re not. Because it’s all about how children learn, it’s not a quick fix,” she says.

Clarke, who is a former primary teacher and lecturer, who now delivers teacher training, has carved out a career researching and refining formative assessment practice and distilling her findings for schools. 

So, what has this work taught her about where schools go wrong with formative assessment, and how they can get it right? Tes sat down with her to find out.

Formative assessment is something that every teacher probably thinks they already understand. How would you define it?

It is a set of teaching and learning strategies and processes that enable learning, rather than measure it. 

There are several component parts to it. The foundation is that there is a learning culture, where pupils have self-efficacy and know how to learn, and teachers have high expectations and a belief that all pupils can succeed. 

There’s pupil involvement at the planning stage, discussion of what excellence looks like and effective questioning. 

Within lessons, importance is placed on pupils knowing learning objectives and co-constructing success criteria. There’s also a focus on talk and on effective self, peer and teacher feedback.

That’s a lot more extensive than the simple “assessment for learning” definition that many would probably have in mind. Do you think the concept is often misunderstood?

Yes. American psychologist Carol Dweck has the same problem; fixed and growth mindset is known worldwide, and completely misunderstood. It must be like when celebrities get misquoted. The minute you become famous, things become garbled and distorted.

I also think the label “formative assessment” brings about misconceptions. 

In the early 2000s, all of the people involved in formative assessment realised that it should have never been called “formative assessment”. People misunderstand it because of that label.

In what way?

There’s an idea that formative assessment is something you do to students. But it’s not. And it is not a test, either. That is summative assessment. Formative assessment is much more complex. 

In America, for example, teachers often see formative assessment as giving students a test, then looking at all the information and using it formatively. That might be a useful thing to do, but that’s not the conceptual framework of formative assessment. 

What exactly does the conceptual framework involve in practice?

It involves a continual quest to understand what children are understanding. This is about questioning, learning stops and walking around the room and getting and giving feedback constantly. 

Don’t sit back and let them get on with their work quietly. Always be on the move. This means you can get there in the “golden moment” to point out tiny things, or have a two-minute discussion. Feedback is too late if you take the books in and realise that ten children haven’t understood something. You’re moving on to something different tomorrow, so when will you have a chance to rectify it?

The balance between in-lesson and post-lesson feedback has massively shifted in the last 20 years. In-lesson feedback used to be very minimal, but we’ve now reversed that. 

What’s happened over the last few years is that teachers are finding that a lot of the marking has been done during lessons, and there’s far less to do after lessons. 

Do you think that is a positive shift?

I do. If teachers spend all their evenings marking, what impact does that have on the learning, compared with putting the effort into a couple of hours of really good planning and resourcing? As educator John Hattie says: “Know thy impact.”

A school in Oxfordshire I once visited asked teachers two questions: “Why are we doing it?” and “What is its impact?” For anything to become school policy, it had to answer those questions - and the reason for doing it couldn’t be “Ofsted” or “the parents”.

If the majority of feedback is now being delivered in class, that has implications for how teachers plan and structure their lessons. What do they need to consider here? 

They need to think about the fact that there are different methods of delivering in-lesson feedback. There’s not just the on-the-move feedback with pen in hand, there’s also the mid-lesson learning stops with the visualiser, where one child’s work is put under the visualiser, and the class looks at their successes and where they could improve. This really consolidates their learning.

There’s also cooperative feedback: when two or three children, at the end of a lesson, or towards the end of an assignment, read out their work to one another. Their peers then comment on what they like and where they think improvements could be made.

Beyond thinking about opportunities for feedback, teachers also need to embed learning intentions and success criteria. These are structures that support formative assessment because they help learning. 

How can teachers make the most of success criteria?

The key point to remember is that while success criteria offer a breakdown of the learning intention, it’s not about what success looks like at the end, but the processes involved in achieving the learning intention. 

When pupils make a bowl in design and technology, for example, the criteria should not be “make a bowl that can hold water”. Instead, it should list the steps involved in making the bowl.  

This breaks down for different subjects in different ways. With children’s writing, for example, you can’t say: if you do these six things, you’ll have written a beautiful piece of writing. It doesn’t work like that. Instead, we need to approach success criteria as tools in a toolkit. 

You can provide students with good suggestions of how you could make your writing full of suspense, for instance, but it won’t guarantee you do an excellent piece of writing. 

What approach would be more likely to guarantee that?

Only exposure to and analysis of excellent texts leads to great writing. When writing prose at any age, children really need to see what excellence looks like before they start to write for themselves. 

I always say to teachers: imagine, when you were at university, if instead of being told to write a 10,000-word assignment, you had a chance to see previous people’s assignments, both excellent and mediocre, to compare and analyse them, to see what features the good ones had that the mediocre ones didn’t. Not only that but what if you’d been given the criteria for how you were going to be marked and assessed? Everybody would get an A. 

I remember doing my master’s degree, and I didn’t even get given a title; you were just thrown into the deep end with no idea of how you were going to be assessed. That is how children can feel if you don’t show them what excellence looks like.

As well as shared learning intentions and success criteria, facilitating children’s dialogue in the classroom also matters. Dylan Wiliam was doing work on this in the late 90s, explaining how we need to be activating children as learning resources for each other. 

How can teachers do that?

In the early 2000s, my teaching teams and I looked at the best way of partnering children so they talk to, and learn from, each other. 

At the time, it had become well established that we weren’t giving children long enough to answer questions, and so, when a teacher asked the class a question, they’d say: talk to the person next to you. 

Doing my own research, I found out that sometimes children would sit next to the same person for six or eight weeks at a time. I believe that children need to have a wide range of both social and cognitive learning experiences to be able to learn from each other, changing who they sit next to on a regular basis. This was not generally happening at the time in most schools.

So, I asked my teams to experiment with different ways of pairing children and to think about how long children should spend in those pairs. They all agreed on the best approach in primary schools: random talk partners, sometimes groups of three, which changed weekly. 

Why is it important that the pairs are random and change regularly?

If it’s random, you might have the highest achiever with the lowest achiever and, for one week, they both get a fantastic amount out of it. However, after that, it starts to dilute. 

But with random talk partners changing weekly, you get mixed abilities, and it cuts through cliques and bullying because children are making new friends. 

At secondary level, you can change these every four or five lessons because, obviously, the children aren’t together as often.

To return to your framework, you mentioned that self-efficacy is the foundation of it. Why is that?

Self-efficacy is your belief in your ability to achieve.

It seems as if, in schools, we deliberately try to lower children’s self-efficacy by putting them into competitive comparative situations. These things carry on in schools, despite 40-odd years of research to show that they’re actually harming children’s self-efficacy. 

There’s an invisibility effect at play here. Children don’t put their hand up and say: excuse me, you’ve totally demoralised me by doing this. So teachers don’t see the real impact. 

There are a lot of barriers to children’s self-efficacy. 

What are they?

Being ashamed of error is a huge one. We need to normalise error in the classroom. Teachers need to make clear that when you’re struggling and thinking really hard, that’s when the neurons are connecting - that’s when the learning happens. If you’re not thinking about something, there’s no way you’re ever going to get it into your long-term memory. 

Praise is another barrier. All the research shows that if you make praise ego-related, and say things like “clever boy” and “clever girl”, it has a worse effect than no praise at all.  

Instead, we need to give task-related praise - “Well done, I can see you’ve managed to do this” - and learning behaviour praise. Lots of teachers give ego-related praise subconsciously; pointing this out to them and discouraging the use of it makes such a difference in the classroom. 

The barrier I battle against the most, though, is rewards. In so many schools, there are still rewards. 

Why are rewards problematic?

There is a lot of research which shows that comparative rewards - where you give five children a sticker and not the others - don’t do anyone any good. 

Rewards that everyone gets are fine, but comparative rewards lead to complacency and demoralisation. This is often invisible to teachers. So many schools give these rewards, and they have no idea of the damage it’s causing.

Meanwhile, we’ve become complacent about what the real reward is: education. Learning is the reward. When did we get to the point where instead of saying: “Oh, look, you’re able to write”, we now say: “Well done, and here’s a patronising token to stick on your jumper”? 

I run learning teams with 30 teachers from 15 schools. We meet three times a year and I work with them to embed formative assessment into their schools. I make it compulsory that they get rid of all their rewards. 

As well as ditching rewards, schools need to get rid of all ability grouping for formative assessment to really work. 

Why is that so important? 

Because ability grouping also damages self-efficacy. Once you are placed in an ability group, that becomes who you are; it’s a self-fulfilling prophecy. 

All of this comes back to self-efficacy. That’s the starting point, and without it, I question how much the different elements of the formative assessment framework will be effective. When children don’t have self-efficacy, they just shut down.

Is there anything else schools need to put in place to make the framework effective?

There needs to be a shared understanding of what we mean by “learning”.

I talk to teachers about a set of “learning powers”, which are things like concentrating, not giving up, being curious, always improving. These powers should be embedded throughout the entire school; everyone should have the same language and way of talking about what learning actually is.

If you do that, and also normalise error by telling children that learning should feel hard, you can create a culture where everyone is in it together. 

Competition, however, has no place in formative assessment, and it should have no place in schools. The motto I give is: “There’s only one person you should be in competition with, and that’s yourself.”

It’s interesting because that’s what Ofsted is looking for. They’re looking for progress for every child. That’s also what formative assessment is about.

You’re now working on helping teachers to harness the principles of formative assessment to support children’s mental health. Can you tell us about that?

I believe the links between formative assessment and mental health are very strong: the self-efficacy strategies, knowing structures and success criteria, easing the cognitive load, and the physical presence of on-the-move feedback. It can all help to reduce anxiety.  

It also leads to a healthier work-life balance for teachers. Simply by moving to in-class feedback and reducing after-school “deep” marking, teachers have more after-school time.

So many are faced with unmanageable and often unnecessary admin or marking. Their mental health needs protecting, too.

What is the one key thing that you want more teachers to understand about your work?

Formative assessment is not a quick fix. It takes a lot of deep time and discussion to embed.

It’s no good saying: well, we’ll keep the rewards and the ability groups, but we’ll use the learning intentions and success criteria and everything will be fine. No, it won’t. Everything links. 

I understand why people are sceptical at first; doing formative assessment well means making major changes to the school. But when done with real thought, reflection and dedication, it is transformative. 

Shirley Clarke was talking to Kate Parker. You can find out more about her publications, current courses, learning teams and online sessions via her website. You can also follow her on Twitter @shirleyclarke_