Chalk & Change 3: Daisy Christodoulou – I’m saying knowledge is good

“In 2007, there was this total monoculture of bad ideas.”

In this episode, we talk to Daisy Christodoulou, Director of Education at No More Marking. Daisy was at the forefront of a movement of bloggers and thinkers show sought to change how teachers thought about student learning, and what they did in the classroom. Her 2013 book, Seven Myths about Education, contrasted good practice – as described by school inspection reports – with the evidence around how people learn. Daisy trained with Teach First, worked at Ark Schools on curriculum and assessment design, then moved to No More Marking, where she’s working to make assessment faster, more accurate and more useful. She recently published her fourth book, I can’t stop thinking about VAR, which applies her wisdom about assessment to the football field. Her current writing is on the No More Marking Substack.

We discussed:

• Why she wrote Seven Myths: what Ofsted reports showed about perceptions of effective practice in the early 2000s, why finger puppets aren’t a great way to teach Romeo and Juliet, and the initial reception the book had
• Her role at ARK Schools and what King Solomon Academy was like
• Michael Gove as a Maoist, and his lasting significance for English schools
• The academies programme, the shifting role of trusts, and whether it would ever be possible to unwind academisation
• Her advice for countries trying to improve their school systems

Daisy was as thoughtful, entertaining and erudite as ever, but I particularly enjoyed her fair-mindedness, as she jumped to offer both argument and counter-argument unprompted.

You can listen to the episode on Spotify, or Apple Podcasts, or read the full transcript below, lightly edited for clarity.

Tell us a bit about yourself. Who are you? What do you do? What should listeners know about you?

I’m the Director of Education at No More Marking. We are a provider of online writing assessments, using a technique called comparative judgement, which is an innovative and interesting way of assessing. We work with about 2,000 schools in England, Australia, the USA – and all kinds of places – on that. I used to teach English. I’ve worked for a big academy trust in the UK. My big things are assessment and curriculum.

You’ve written four books, Seven Myths about Education, Making Good Progress?, Teachers vs Tech? and I can’t stop thinking about VAR. If they are in the unlikely position of not having read any of them, what should listeners read first?

If education’s their thing, Seven Myths about Education would be a good place to start. That was the first book I wrote. It’s about ideas that have traction within education, but don’t have much evidence behind them. I wrote it back in 2013 – it came out in 2014 – and it’s still very relevant, even though a lot of things have changed since then.

If you’re into sport, you should read my book about VAR, which is very different. It’s about the Video Assistant Referee that’s used in football. It explains what some of the problems are with it, applying some assessment and measurement theory. It’s a departure from my day job, but there is a link in that refereeing is analogous to assessment.

You’ve written those four books while working full-time. What’s the secret? What’s your working and writing process?

I don’t know. I try and make the most of every minute – even if you only have 15 or 20 minutes spare, you can often get something written in that time. That’s how I like to operate. If you always wait for there to be a clear half a day, you’ll never get anything done. I find you have to fit stuff in, around – even if you only have 15 or 20 minutes, you can get something done with that time.

Let’s start in broad-brush strokes. How have you seen English schools change over the last 20-odd years?

I trained to teach in 2007, which feels like a really long time ago now – culturally it’s quite a long time ago. The summer I trained, Tony Blair was standing down as Prime Minister, and I remember, when I was on a residential training course that summer, we were watching the rolling news about him leaving office. It was a very long time ago.

What’s changed in schools in that time – in Seven Myths, I talk about some of these very pervasive ideas about education that don’t have much evidence behind them. When I trained to teach, they were the dominant culture.

Seven Myths has got seven myths in it, as the name might suggest, but the underlying argument I’m making is that people are very negative about teaching knowledge – they think you don’t have to teach knowledge in the modern world. What I was saying in the book is, “If you look at all the research evidence from cognitive psychology, knowledge matters. It doesn’t matter that you have Google to look things up. You still need knowledge and long-term memory.” That’s the basic thesis of my book: everyone says knowledge is bad. I’m saying it’s good.

You’d never written a book before. What made you think, “This is a thing and this needs to be a book”?

When I trained, and in my first few years of teaching, there was some really far-out stuff, that I was being told was best practice, and “This was the way you had to teach.” Loads of it just seemed to offend common sense. Then I started digging into the research and it was, “This doesn’t just offend common sense – it directly contradicts so much research.” I felt I should try and do something about this, try and communicate more about it.

In 2007, there was this total monoculture of bad ideas. Now, it’s not like that anymore. These ideas that I was talking about in Seven Myths are now much better known. “Knowledge” it’s not the nasty word that it was in 2007. People are much more aware of the way that knowledge is important if you want to develop skills. There’s a much better understanding of the basic psychology of learning, and the culture’s really shifted. That’s been a big change since 2007 – for the better, I should be clear.

People are publishing books all the time, and good ideas get unfairly neglected. How would you explain those ideas being picked up and becoming mainstream? They could just as well be picked up in America or Australia. England seems to be a bit of an outlier for this. What do you think happened?

People often ask this. You can come up with plausible reasons, but it’s hard to say, “Was there one decisive thing?” A few key things – a really important book, which was published in 2008 in America, was Professor Daniel Willingham‘s, Why do students like school? He is a cognitive psychologist, from a top global university and he was putting together all these arguments in a persuasive and easy-to-read way – about a lot of the things I was talking about in my book too, about the value of knowledge.

I lean very heavily on his book. An American friend was saying to me recently how interesting it is that that book was published in America, but it seems to have had more traction and influence in the UK. You could almost say the same about E.D. Hirsch. He is very well known in the US, but you could argue he’s had more of an impact on the curriculum in England. Lots of very powerful ideas, being done by experts in their field, are having more traction in the UK than in any other country’s education systems. Why did that happen?

A couple of things happened in around 2010-12. You had a grassroots movement of teachers – some, but not all of whom, have come through the Teach First training programme, like I had – who were questioning the orthodoxies and saying, “Hang on, does this really makes sense?” Then also you had a top-down influence too. From 2010 to 2014, Michael Gove was the Secretary of State for Education [his Chalk & Change interview will be released in the new year]. From 2010 to 2023, on and off, Nick Gibb was the school’s minister [Chalk & Change has interviewed Nick Gibb]. They were very open to and interested in these ideas, and pushed these ideas. That gave them more of a reach than if it had been bottom-up. You had this very interesting coincidence of a bottom-up grassroots movement and a top-down political movement coming together at the same time.

Even though Gove and Gibb are gone, the influence of that is still living on – you can’t put the genie back in the bottle. These ideas are out there. People are talking about them, and other countries are now interested in what has happened in England in the last 15 years.

What was the reaction like in 2013 or 2014?

Seven Myths came out as an e-book in 2013, then Routledge published it as a hard copy in 2014. I got a lot of criticism, especially when it came out as an e-book, because I got people being, “What is this self-published madness?” It’s hard to remember now, because these ideas have become more accepted, but, to put it bluntly, I got a lot of grief. A lot of people saying, “This is crazy. You want to indoctrinate kids. You want to do chalk and talk. You want to take us back to the 19th century. Haven’t you heard about the internet?” A lot of criticism, which you still get a bit of now, but things are more mainstream. It was a controversial book for sure.

You drew heavily on Ofsted reports. Why was that?

Before I wrote the book, I would be having discussions with people about these issues on Twitter. What I used to run up against, again and again, was this weird intellectual position. I would say, “There are schools who think you can teach English through the medium of finger puppets. This is a bad idea. I don’t think you can teach English through the medium of finger puppets.”

I would get this response from the same person, often in the same breath. First of all, they’d say, “Don’t be stupid. Nobody’s planning to teach English through the medium of finger puppets. You are just exaggerating. You are being a bad-faith actor. You are making stuff up that doesn’t happen. How outrageous for you to suggest that schools are teaching through the medium of finger puppets.”

I’d say, “Here’s some evidence that schools are teaching through the medium of finger puppets.” The same person would say, “Teaching through finger puppets is a great idea. Actually we should have more finger puppets.” I’d be, “Make up your mind. Is it, ‘Nobody teaches with finger puppets’ and I’ve created a straw man? Or is it, ‘That straw man is real and it’s a good thing. We should have more finger puppets.'”

What I felt I had to do when I wrote the book was prove that the finger puppets – which is an example in the book – were real. A large chunk of the book is me saying, “I’m not making this up. This is recommended best practice. You can defend it as recommended best practice. Or you can agree with me, it’s a bad idea. What you can’t do is pretend it’s not happening anywhere.

The best place to prove it was recommended best practice was Ofsted, the school inspectorate. They inspect thousands of lessons a year, and produced, and still produce, all these reports where they say what they think best practice is. My whole point was, if the national schools inspectorate are churning out reports saying, “A, B, and C are good practice and X, Y, and Z are bad practice.” I have clearly not created a straw man.

My whole thing was to identify examples in Ofsted reports of what they were saying is best practice, and then to explain why they weren’t. One of the examples is that finger puppet example. The reason I picked that is because that was very controversial. It still is – I still get intakes of breath when people hear that one. Ofsted, in a report – this would be in about 2010 – gave an example of best practice in a secondary English lesson, that was teaching Romeo and Juliet by making finger puppets. I give chapter and verse about why that’s not actually best practice. It was one of those that was particularly controversial and I still feel people want to argue about that.

Do they want to say that it’s good practice?

My argument against – this is almost directly lifted from Dan Willingham – he has this great line, “Memory is the residue of thought. We remember what we think about.” He says, because of that psychological principle, when you are teaching a lesson, you have to be thinking all the time about what your students are thinking, because what they’re thinking about is what they remember. They won’t remember what they want to remember – or what you want them to remember. They will remember what they think about.

This seems like a simple principle. Here’s what’s interesting about it: when you say that to people in the abstract, they go, “Obvious, duh.” But it then suddenly has all these implications, which people are often less willing to accept, and can get quite angry about. Willingham gives an example of a history teacher, teaching a class about the Spanish Civil War. He takes them to the computer lab, says, “Research the Spanish Civil War and create some PowerPoint slides about it.” Within a few minutes, the kids will create animations of the “Spanish Civil War” letters flying into the PowerPoint. They’re thinking about esoteric features of PowerPoint, not the Spanish Civil War. That is what they will remember.

My example is the same. What are you thinking about when you are making finger puppets? The mechanics of making the finger puppet – not Romeo and Juliet. If the purpose of that lesson had been to learn how to create a finger puppet, it’s a good activity. If the purpose of the lesson is to learn about Romeo and Juliet, it’s not.

What’s the counterargument when you say that to people?

The counter-argument is, “You need to motivate students. You need to be doing something that’s fun. If you go in with Romeo and Juliet to begin with, that’s hard and dry. The finger puppets are fun and engaging. The kids really enjoy it. Maybe they’re spending time on the finger puppet, but once they’ve made it, then they can have the finger puppets talking to each other and doing language.” I’ve heard all the arguments. “If you just do Shakespeare, the language is boring, the kids are going to find it hard to engage with that. Most of the arguments you get are some form of engagement.

Do you remember having these insights and changing your own teaching – from what you’d been trained in originally – to apply these more effective approaches?

I remember reading Why don’t students like school? while I was still a teacher. He says, “The most valuable single piece of advice cognitive science can give teachers is, review every lesson plan in terms of what the students are thinking about.” I remember thinking – this wasn’t in my training year, it was later on – if he’s saying that’s the number one thing, let’s give it a go. It was interesting: I wasn’t necessarily doing finger puppet stuff, but I did realise that I was setting up quite complicated activities. You would spend time explaining and getting the kids to understand. That would soak up a lot of time and thought. It made me think – you need to strip things back and get students thinking about what you want them to be thinking about as soon as possible in the lesson.

It is a very useful principle, and it’s useful for all kinds of training and teaching – what is it you want people to be thinking about? It’s like a lot of the best bits of advice: it’s deceptively simple. You hear it and go, duh, obviously I’ve done that. Then when you look at it, you go, “But am I doing that?” It’s harder to do than you might think.

So I did start applying some of these things early on. I did feel, in that period of 2007-2008, there were things that were being recommended at the best practice that, even if you hadn’t read the cognitive psychology, felt so counter to any common sense. You felt you had to push back. There was a whole thing around, “Teachers shouldn’t talk for more than two seconds [I assume Daisy means “minutes” here].” You get these crazy Ofsted consultants coming around and saying, “If you’re talking for too long, the students aren’t being active enough.” You would think, “How is it possible to manage a classroom?” If you’d ask people, “What does this look like?” they would never have a good example for you, because it is exceptionally hard to run a class of 30 14-year-olds – even if you’re not bothered about teaching them anything – without speaking for more than two seconds, this is difficult stuff.

I found my checklist for my final assessed lesson at the end of my initial training, and it included an expectation that you differentiated according to learning styles.

Learning styles is a great example. Even at the time, people were starting to push back about it. They have never had any evidence behind them, but they’re very seductive and people really seem to buy into them. I remember those being a huge thing when I trained too. Not just that you would have the kids’ learning styles, but I remember that we would be saying, “As a teacher, what’s your learning style? You can learn better in that style too.” That was a very powerful idea at the time as well.

Talking about the influence process, you mentioned bloggers and mentioned politicians. Would you add anything – in terms of those ideas being embodied in the new curriculum, in terms of changes in inspection, or whether academisation gave them wings?

This is an interesting question. You’re talking about, “What’s the relationship between structures and ideas?” You had a bunch of powerful new ideas that were being kicked around, both in America and in the UK, in about 2009-2011. How do you turn those ideas into practice? What is the role of school structures and incentives? Then the debate is – academisation: does that help these ideas go further?

When I started teaching, you had all these bad ideas around, but you had a lot of academies. I started teaching in a new academy. Academies weren’t brought in by Michael Gove with the Conservative government. They were introduced by the New Labour government. Gove put them on steroids and created more of them. They have more freedom from central government, and so potentially the freedom to think differently and do different things on the curriculum and other things as well, like teacher pay and conditions.

To what extent did they help these ideas spread? You can make the case either way. You can make a strong argument that they did help these ideas spread. The argument you can make for that is if you look now at some of the best schools in England, in terms of the new Progress 8 measure, a lot of them are schools in academy chains, and a lot of them have embraced these ideas. That’s a powerful argument for academisation – helping these ideas spread.

I don’t necessarily think it’s an open and shut case though. If you want to make the argument against, when I started teaching, there were academies and nobody had heard of these ideas. It can’t be structures. I often like to talk about the difference between Russia and Estonia after the Soviet Union breaks up. They have very different pathways. The Soviet Union collapses and you have a lot of new ideas and new structures. But Estonia takes a pathway of free markets and democracy. Russia doesn’t really. Things can’t be about structures. I don’t know, it’s a tricky one. I don’t want to be blase and say structures don’t matter – they obviously do. But it’s really interesting that in Britain in 2010, you had this coincidence of something bottom-up and top-down. You can have top-down structural changes and they can change the incentives in the system, but where do those ideas come from? You have to have the right ideas, and people prepared to think about them, believe in them, and do things differently.

You’d spent some time in teaching. You then went to Ark for a number of years. What made you move from one school to a trust?

Ark was one of the first-ever, academy chains. I started to work there from 2013 – at that point, they’d already done lots of fascinating things. King Solomon Academy, which was one of the pioneering new academies, was an Ark school. They were one of the most innovative, thoughtful academy chains, as they still are. My job originally was director of research and development. At that stage Ark had about fourteen schools, primaries and secondaries. It was a great opportunity to get these ideas to a bigger audience, and in a place where they could make a practical difference. That was how it proved, it was a good time there.

What did you see Ark doing that other schools or trusts weren’t doing at that stage?

Look at how the academy movement in England has shifted. When the academy movement began, Ark was one of the first groups, and was setting up some of the first academies. When the academy movement first began, it was about trying to create amazing schools in places where historically there’ve been struggling schools. It was very much about being in disadvantaged communities, in tough places where – perhaps, it hadn’t been that a school had been failing for ten or twenty years, but for decades. The original academies movement under New Labour had been about, “Let’s go in and try and do something differently.”

It was influenced by that American movement. The American equivalent of academies is charter schools. The reason they’re called charter schools is you would have a charter for ten years to do something differently. Because these were schools that had been underperforming for a very long time – why not go in and try something different? That was the original thesis behind academies, and there weren’t very many of them as a result. They were in, deliberately, tough areas.

Then Michael Gove came in, changed the whole idea behind academies and said, “Any school can become an academy. You can become free of local authority control, be funded straight from central government, and have more freedom.” An incredible number of secondaries took that opportunity. Not as many primaries, but everyone was surprised by the uptake of secondaries saying, “We’re going to do this.” Then the academy movement is something very different.

I was at Ark as that shift was happening. I joined in September 2013 – three years after Gove published a white paper and let schools have these freedoms. You were going from a world where academies were this niche, special case to – when I arrived at a 2013, people were saying “Multi-Academy Trust (MAT).” MATs were not talked about under Tony Blair – that was not even a phrase. I’d love to do the Google Ngram on when people start talking about “Multi-Academy Trusts,” because it wasn’t a thing.

It was probably when I started at Ark that people started saying, “MATs are now a major structural feature of the English school landscape. They’re going to have a massive role to play.” You’re saying, “Maybe every school is going to be in a MAT.” That was a huge shift that we were all living through when I was at Ark. In some ways, everyone is still living through it. Now, fifteen years on, you’ve got Bridget Phillipson maybe trying to unpick that settlement. Who knows what that looks like.

Ark has had to evolve and go with those changes, as have all of those early academies. You’ve got a lot of other MATs that have come into being much later and have been more specific about what they’re doing.

In terms of how that affected things when I was at Ark – when it began, it was very much about trying to do something innovative and perhaps having a handful of schools that were doing something really different. You would look at King Solomon Academy – it had these phenomenal results and was an absolute flagship new school that set up in a difficult area and got incredible results. That was the early wave.

Then you are moving to the system you’ve got now, where the point of a MAT is much more to have standardised structures and routines in a central office, that can be rolled out to multiple schools. New schools can come in and every school follows the same script. That’s been the shift in the idea of academies over the last fifteen years or so. They’ve gone from being this one-off, try something out, in places where things are failed, to being “You are now a structure that has to look after lots of schools. You have a central model, and you roll that out across quite a lot of schools.” That’s been a big shift. I don’t think that was ever even necessarily a willed shift. It’s almost happened by accident.

To oversimplify, do you think that’s worked?

If you want the critique – all this has happened because of Michael Gove. He came in in 2010, and he really blew a lot of stuff up. If you want the critique of Michael Gove, he’s not a communist, he’s a Maoist. He blew up the established system of local government oversight of schools that had been around since 1944 – and had its roots in stuff that went way beyond 1944. He blew that up, because the finances were dependent on the secondary Local Authority (LA) schools – having a chunk of their money going to the LA. Once those secondaries peeled off, the LAs didn’t have the money to maintain the stuff that they did.

Do you think that was a Machiavellian move?

Yes, I think so. That was the point. Gove didn’t have much faith in the local authorities doing a great job for education. The secondaries spin off and take their money with them. The local authorities don’t have that money to do things anymore. Gove didn’t think they were doing a great job: you can say, “We’ve reduced their negative impact.”

If you want the critique, he blew something up without knowing what would replace it. The thing was always – you get these big secondaries, with big budgets, a lot of students and teachers, and they do have the capacity to go alone. Most primaries are one-and-a-half-form entry maybe. They have a head teacher who maybe is teaching a day a week. Most primaries don’t really have that capacity. You’ve created a system where they probably do need to have to be in groups. At primary, it probably does make sense that those groups are geographical. Everyone always thinks about schools in the context of secondaries, but there are 16,000 primaries and only 3,500 secondaries in England.

So it blew up a system that, whether you liked the ideas that were coming out of them or not – and I would tend to agree with Gove, I probably didn’t – it did a lot of mundane things, like risk management and risk pooling, HR and payroll, and if a kid needs a hearing aid, you’re not going to go to Whitehall to get a one. You want to go to a local government. These basic, boring functions that are just really important. A lot of that got blown up. You’re in a system now where the Department for Education directly funds an awful lot of schools centrally. For all that governments have tried to encourage schools to join MATs, and not just be standalone academies – one of my favourite stats is the modal number of schools in an academy trust is, I think, still one. The median number is still two. It’s quite hard to run a big system like that.

The system is very messy. Is messiness a bad thing? Gove was prepared to have and tolerate the messiness. I find it really interesting though – he is probably the most consequential education minister. People talk about Butler, quite rightly, the Butler Act 1944 and then the Baker Act. Then there’s all the getting rid of grammars – Thatcher shuts more grammars than anyone else. But Gove, in terms of the structures – forget about the ideas for a minute – it’s completely consequential.

You’ve seen with Bridget Phillipson – she’s trying to turn the clock back. But it’s now really difficult. If you run some alternate universe where the Tories get kicked out in 2011, and a Labour government come in and Bridget Phillipson is in Education – at that point you could’ve put the train back in the station. You’re now fifteen years on. The train’s left the station. All this school improvement capacity is in MATs, not LA’. How do you reverse that and put everything back into local authorities? Even if you think that’s a good idea – I probably don’t – but it’s just hard to do.

Gove is enormously consequential, because you’ve got someone coming in, deliberately trying to undo what he did, and finding it really hard. I don’t see any way that – whatever you try to do – you resolve some of the decisions Gove made. We’re going to be living with them for many years to come. Even if you want to put all of the schools back in an LA, or – Nadhim Zahawi was trying to do this – take the next stage of the Govite revolution: get every school in a MAT. Both of those things are really quite hard to do. If you want to do either, and you absolutely threw your weight behind it, and were a politician of genius, and had money to spend, which nobody does – I still can’t see how you would do it in under five or ten years.

You’d be talking about almost a quarter-century from when Gove came in, and you’d still be living with the consequences of that white paper in 2010. We’ll be living with it for much longer than that. He set the contours for the next literally half century.

It’s great that you mentioned King Solomon Academy (KSA). What was it like? What were you seeing there?

When I started at Ark, KSA had this, quite rightly, incredible reputation. We were all in awe of it. If you were joining Ark new in 2013, you were like, “What do they do? This is an amazing school, it’s got these unbelievable results.” I’m not taking any credit: once I joined there, it was up and running. I came in the wave after that, and got to know a lot of the people there: Max Haimendorf, Natasha Porter, Bruno Reddy, some of those people who helped set it up.

It was doing things which then were very pioneering and in some ways quite shocking, that now have become standard. One of those would be family lunch. Those guys went and visited a lot of the American charters. The idea was they’d all sit and eat lunch together. Now it seems so obvious, so many schools do that. At the time it was a really big deal. They were doing things, coming up with these ideas, which were then spread across the board.

I remember first hearing about KSA at a Teach First event, round about when it was being set up. There was a huge thing about it being a small school. I remember one of the people involved talking about human-scale schools. It was two-form entry, and that was going to be a really big thing There’s fewer people for everyone to get to know, so you can have more human scale.

I really bought into that as an idea – I kind of still am bought into it in principle – but as I’ve been around in education and worked for and with trusts, you also realize how hard it is, first of all, to scale it up. Small schools are more expensive. Also, it’s not just economies of scale – you often can’t offer the same range of subjects, particularly when you get into the Sixth Form, with sixty kids in a year group.

There has been an interesting tension between the new school/transition school model. I love the idea of, you have new schools doing something innovative, and they can come up with really interesting things – like the family lunch – that you can bring in. A new school makes it easier to do that. But you also have to realise that most schools are never going to be new schools. What do you do about established schools that are six- or eight-form entry? It was easy – too easy and unfair in ways – for people to turn around with King Solomon and say, “They can do all that ’cause it’s just two-form entry. We’re eight-form entry. We could never do that.” That was too glib, but there was some truth to it.

That’s been one of the interesting tensions about all of the new schools. Michaela Community School would be an example. New school set up from scratch. Very obvious clear vision. People would say, “You can do that in a new school, but what about a school that doesn’t have that ability to set up from scratch?” But, in defence of the King Solomons, the Michaelas, all these schools, they have brought in so many ideas that a lot of established schools have been able to copy.

It is hard copy everything that a school does. But there are so many things you can look at and go, “That’s something we can do.” There are a lot of schools doing that. To go back to King Solomon, it was fantastic. It was great to be able to go in and see such an amazing school close up and to see what those guys were doing.

You only need the twenty, forty, fifty schools that are those real role models. You don’t need to make every school like them, for them to have that huge effect.

This is it. People have a big debate about scale. I think about scale a lot, because I work now in technology. We are trying to assess hundreds of thousands, millions of pieces of writing. Your critique of these small, one-off schools is, “They’re not scalable.” People will say, “Even the biggest Multi-Academy Trusts don’t have that many schools. If we really want to make an impact, we need MATs with a couple of hundred, maybe a couple of thousand, schools. What is the point of one school when you’re talking about a system that has 20,000?” I do have some sympathy with that argument.

I do have some sympathy for the argument that, if you want to make one school work, you do stuff that works for that one school and is not replicable. Some of these schools have been real beacons and have shown what is possible. They have enabled scaling of a different type. They’ve got high profiles. People have been able to look to them to see how different ideas work in practice. I am passionate about the idea that we do need to think about scalability – we can’t just be doing things in a bespoke way that only work in a certain context. But the handful of amazing one-off academies you’ve got have had an impact beyond their doors, not just on the couple of hundred kids who go to that school.

We’ve talked in general about the trust. Tell me about what you were doing for Ark. You were director of research initially, and then you were director of assessment. What were you trying to achieve there?

I was director of research and development for two years, and a lot of my job – it doesn’t sound like it – involved working on a pioneering English Mastery curriculum, which goes on to this day. Then I ended up getting involved in assessment. As trusts get bigger, having reliable assessment data becomes more of a challenge and more important. The government had got rid of National Curriculum levels. I moved to be Director of Assessment for two years, and worked on various things around getting assessment right and getting that central data in a good place. That was fascinating and led me down a lot of assessment rabbit holes. That was how I then ended up working for No More Marking, where I’ve been ever since, working on exclusively on assessment.

We’ve talked about trusts as improvement mechanisms and that now the main thing we want them to do is help schools improve rather than just set up new schools. Any reflections from that time – trying to work with fourteen or fifteen schools, as it was – about how easy it is for a trust to change things?

It’s very hard. I was at Ark at an interesting time, when the system was shifting from this idea of academies being, “Let a thousand flowers bloom” to a world where we were saying, “We want a central team, who are going to roll out a central model, and a team that’s going to have school improvement capacity to go into a failing school and turn it around.” That’s not easy. Even if you have a bunch of existing successful schools, do you have that model in such a way that you can roll it out?

One of the things you saw at that time – this didn’t happen at Ark, but it did happen – there was a big push from central government for MATs to scale up. Two of the biggest ones that did scale up ran into big financial and educational issues [E-ACT and Academies Enterprise Trust]. Ark and Harris were interesting, because they were high-profile early academy chains who didn’t scale up that fast, because they were wary about this and they realised it wasn’t that simple. They came out of in a stronger shape. They only took on what they could take on.

You saw an issue at that time with a lot of schools and trusts trying to grow – maybe thinking they had a model that they could roll over into schools. You’d often end up with the worst of both worlds: existing heads of good schools being taken to support the weaker school that had just come in. The good school that they were now neglecting was not actually strong enough to cope with the loss of leadership. Then that head was being pulled in different directions and you would end up with two schools struggling instead of one good and one struggling school.

A lot of people who’ve been in the system in the past few years will recognise that’s an issue. All these things are about your capacity.

• Do you have a model that you can roll out to a new school that comes on board?
• Is it a robust model?
• How dependent is it on having an excellent headteacher
• If it is dependent on having an excellent headteacher, where is your pipeline of headteachers coming from? Is that a key part of the model?
• Similarly with teachers, how good a model can you have if you haven’t got a pipeline of good teachers coming in? Where are they coming from?

All of these things are pretty hard to do at scale. They’re hard to do in an individual school. But they’re very hard to come up with a model. You’ve seen, as times gone on schools and trusts getting to grip with this more. But it’s not easy.

We’ve talked about ideas, people, political leadership, academies and pedagogy. If the Education Minister of Ruritania comes knocking on your door and asks, “What three things should I do?” What would you take from what’s happened here?

I’m going to be cheeky and say what is more important depends on what stage you are at. There’s a point at which ideas are more important. Then there probably does come a point where structures and incentives are more important. When I started teaching, the ideas were more important, because there were so many bad ideas – you had to get that flow of good ideas into the system. Now, there are lots of good ideas in the English system, and maybe it is more about structures and incentives. How do we reward the schools and MATs who are getting the right thing done and get them to flourish? What do we do about the schools that are still struggling? So it depends on where your system is. If the Education Minister of Ruritania comes along – I’d want to look at Ruritania and say, “Where are things at the minute?”

That’s a good teacher’s answer – it depends on prior knowledge. Are there any factors we’ve not talked about that you think are important?

Phonics is a really interesting example of something that worked quite well. As I said, some of the stuff Gove did blew things up. I’m someone who thinks it was more good than bad, but it was chaotic. Phonics is an interesting case study of something very specific, quite niche, but still very important, where lots of nice policy elements aligned really well. For example, having matched funding for schools, so they could pick their phonics program – there was still a choice of programs. It didn’t turn into a monolithic government provider that wasn’t incentivised to have good production values. The phonics check was really good.

Another aspect of the school accountability structure we haven’t necessarily talked about is – Progress 8 was really good. Quite a wonky, technocratic change addressed some of the issues with previous accountability metrics. Lots of assessment changes – getting rid of coursework, which now looks a real stroke of genius, given ChatGPT and the impact AI plagiarism’s having on everything.

Then Ofsted – I started off with all the issues with Ofsted and all the bad examples in their reports from 2010. But Ofsted went through a very big reform process under Amanda Spielman. [It’s] now potentially going in a different direction again. Ofsted is hugely influential – plays a huge role in the structures and incentives of the system and has been on a real rollercoaster ride. All those things are important as well.

 You’ve mentioned No More Marking, what does it do? Why is it useful?

We provide schools with online comparative judgement assessments. Comparative judgement is an amazing way of assessing extended writing – which we do a lot of in England. The traditional way we assess essays is not very effective. It’s a form of absolute judgement, where you look at one essay and you mark it against a mark scheme. The problem is human beings are not very good at making absolute judgements. We look at one thing, and try and place it on an absolute scale – that’s actually really difficult.

Imagine someone walks into the room, and I ask how tall that person is. It’s quite hard to get it right. Comparative judgement – imagine two people walk into the room and I say, “Who’s taller, the person on the left person on the right?” That’s much easier. We do that with student writing. We say to a teacher, “Here are two pieces of student writing. Read them both, which do you think is the better piece of writing?” They don’t just make one decision – they make a series of decisions. And it isn’t just one teacher making decisions, we run these enormous national projects where we will have thousands of teachers, making hundreds of thousands of decisions, on thousands of pieces of writing. We are then able to construct a measurement scale of all the pieces of writing, and we can link those measurement scales over time.

We have been doing this for nearly ten years. We’ve got a measurement scale that runs over ten years. We can tell you what writing was like in 2017 pre-pandemic – the average score of a Year 6 cohort then – compared with one now. That’s powerful stuff. That’s also hard to do with traditional assessment.

In the last year or two, we’ve started to build in Artificial Intelligence into our systems. Teachers can still judge, but if you want to reduce your workload, cut a bit of time, you can add an AI judge. That’s been interesting because we’ve got to the point where we think the AI judges are pretty reliable. Also they pick up on some human errors – the big one is handwriting bias. Our teacher judges are pretty good, but if they have a weakness, they can often be swayed by better or worse handwriting. The AI is not susceptible to that.

That’s where we are at the moment. We think the combination of comparative judgement and AI is incredibly powerful for giving you more reliable assessment results, getting them to you quicker, and giving you more useful feedback.

If you’ve got data going back ten years, have you seen Year 6 get better?

We’ve seen a bit of plateauing. Immediately after Covid, we saw a massive dip, but we saw students make it up relatively quickly. It’s not learning loss – it’s learning decay. They made it back pretty quickly. We haven’t seen enormous dips. You might think that’s crazy. I did an episode talking to More or Less about this recently. It’s not just us. There’s a lot of good, high-quality data that, if you dig into it, the pandemic effect is not as big as you might think.

The biggest gold-standard finding in this is Ofqual. They run a national sample test every year for 15-year-olds, and this is one of the highest quality, most rigorous things out there. It was deliberately set up to monitor standards over time. It’s the same test delivered every year to a cohort of Year 11s in March. It’s all completely secret and anonymous – and it’s low stakes for the student. You can get a real idea of where the cohort is.

If you look up what Ofqual published about it, it’s been steady, pre- and post-pandemic. You can look at that a couple of ways. You can say, “The pandemic’s had no impact.” You could say, “Maybe if the pandemic hadn’t happened, kids in England were on track to do really well.” There’s some international survey data where England scores have held steady, but their ranking has jumped – because all the other countries have got worse [I’ve reviewed England’s standing in international test results]. There’s a number of different ways you can interpret this, but our data shows – we haven’t seen enormous falls.

The other wrinkle I’d throw into this is, there is an issue with absenteeism, which may not be being picked up in our data, or Ofqual’s data. I don’t know enough about that. But we know absenteeism is a big issue. A potential risk with the quality of data is, are we fully capturing all these students who are absentees? I’m not saying Ofqual are or aren’t. I’m saying that’s a big issue for all assessment data at the minute because we’re dealing with levels of absentees that we’re not used to.

Where can listeners find out more about your work?

Our Substack is the best place – we keep it updated with our latest.

Daisy, Thank you very much.

Thanks Harry. Enjoyed it.