This year about a third of kids didn’t manage to achieve a grade 4 in English and maths at GCSE. That’s about the same as last year, and the year before that, and the year before that. And so this year, like all years, it’s led many to call for Something To Be Done.

There is a forgotten third, the say, who are leaving school feeling like failures, and without a crucial piece of paper that unlocks most future opportunities to them, including further study and employment. This wound feels particular sore and the issue particularly urgent given Alan Milburn’s recent interim report into the one million NEETs in England.

Is ‘not a pass’ a ‘fail’?

There’s a few things that we should clear up.

The first is that a grade 1-3 at GCSE is technically a pass. It is classified as a level 1 qualification (whilst grades 4 and above are level 2 certificates). However, calling them passes feels pretty disingenuous to many critics for three reasons:

1. The government officially designated a grade 4 as a ‘standard pass’. It follows for many that if you haven’t got a pass, you’ve failed.

2. There are consequences to achieving below a grade 4, some informal (sixth forms and colleges may not not let you enter, or not let you study that subject at level 3, or at least heavily advise against it) and some formal (for maths and English, pupils are required to continue studying maths and English, but more on that later).

3. In part due to 1 and 2, and in part due to public perception, grades 1—3 psychologically function as a ‘fail’.

The second technical point is about how inevitable this is.

If you plot the number of kids achieving each grade, it looks like a bell curve. A slightly different bell curve depending on the subject, but a bell curve (or normal distribution) nonetheless. This has led many observers to assume that GCSE grades are norm-referenced, meaning the number of children achieving each grade is restricted. Essentially, they believe that each year’s GCSE scripts are ranked from top to bottom and the grades are assigned based on where they are in the pile. This means that for every winner, there is a loser. If you move up to a grade 6, then someone has to move down. This feels inherently unfair. A rigged game. Improvement is an illusion, and some kids, no matter how hard they or their schools work, will always be left with these lower grades and the consequences that follow.

Thankfully, that’s bollocks.1

GCSEs are not norm-referenced. It’s such a persistent and pernicious myth, that the qualifications regulator Ofqual themselves issued an official myth-busting clarification in 2017.

It’s also worth saying that GCSEs and A levels are not norm-referenced either. If they were, we’d see fixed proportions of each grade in each specification, regardless of the ability of the cohort or the ability profile of any one exam board.

It’s regrettable that many commentators who loudly declare that GCSEs are norm-referenced have read this and therefore are aware that it’s not true.

The third and final point of clarification is that the consequences that follow from a grade 1-3 are different for maths and English compared to other subjects. It’s commonly said pupils are required to resit their English and/or maths GCSE if they don’t get a grade 4, which isn’t actually true (there is in fact a condition of funding for their onward institutions that they merely continue studying English and maths, though, for a number of reasons, this means that most kids are entered in GCSE resits, and often quite quickly).2 But it is true that failing to get a grade 4 in English and maths is genuinely consequential: there is a consequence for the kid, and a duty on whoever educates or trains them next.

So it’s all fine then?

When you start to make these clarifications, as I have over the last four or five years, I’m very aware that I quite quickly start to sound at best like a kind of linguistic gymnast, and at worst quite cold-hearted and like a bit of a dick. It seems to be waving away the fact that a third of kids don’t get a grade 4 in English and maths because there is technically an alternative universe in which they do. But that isn’t the universe that we’re living in, and despite all of the reforms and promises and blood sweat and tears the numbers barely seem to be moving at all. It sounds like I’m saying that’s totally acceptable, shifting the blame onto students, their parents and their teachers.

It’s not fine. It’s very, very not fine. And so it’s worth looking at some of the options for what we could do differently so that a third of kids don’t end up with a grade 1-3.

Choose your poison

1. Lower the standard.

As mentioned above, grade 4 is not anchored to a point in the population bell curve, but is instead fixed to a standard which is demonstrated by meeting certain criteria. So if we wanted more kids to get a grade 4, we could simple lower the standard that is required. After all, if a third of kids are consistently not meeting it then surely it’s too high? We could recalibrate so that kids that are getting grades 2 or 3 now would actually be getting 4s or above (and perhaps everyone else would shift up accordingly, or maybe you’d just have more kids in each grade category).

There are four problems with this approach.

• The first is the most serious: explicitly and transparently lowering the standard necessary to pass cheapens everyone’s qualifications, including those who you’ve just rewritten into passing grades. Qualifications have value, partially because of how reliably they signal information about what a person knows and can do, and partially due to the public confidence in them.

• The second is that it changes the label without actually changing the thing. Yes, we care about how many kids are getting a grade 4 or above, but we really care about what that signifies, we want more kids being better readers and writers and mathematicians.

• The third is that the standard expected for a grade 4 is not, in my opinion, an unreasonable one for the vast majority of kids to achieve. Again, I’m aware that this is firmly in the ‘dick’ category, but I think it’s quite hard to look at a foundation maths paper, and conclude that after 11 years of schooling it’s ludicrous to assume that most kids should get more than 60% the questions right.

• The fourth problem is that it’s politically impossible. That may seem morally unacceptable (politicians should do the right thing!) but the reality is that any government who pursued such a policy would face such an onslaught of “dumbing down” attacks that it would die in the water. And, because of 1 and 2, it’s not at all clear that it is the right thing to do, so why jettison your year’s supply of political capital, making it very hard indeed to do anything else?

2. Have fewer grades

We could also go some way to solving this problem by just having fewer grades. Why do we need to categorise children into such fine grained chunks? If we move to five grades, for example, then maybe we retain the bottom grade as a fail, and maybe following the normal distribution logic it would still have roughly a fifth of children in it. But at least a fifth is better than a third.

But just reducing the number of grades, or shifting to pass/fail with current GCSEs has some major issues. Wherever we place grade boundaries, we create cliff edges. The fewer grades you have available, the sharper and more consequential those cliff edges become. If you were two marks of a grade 8 then it’s a bit of a bummer but you still get a grade 7. You’re in the right region.

As you reduce the number of grades, that starts to skew. Now that two marks suddenly puts you in what feels like a totally different category. Like most primary teacehers I’ve struggled to explain this to panicky parents who are told that their children has gone from working at the expected standard to working towards the expected standard. In fact, there are only three ‘grades’ in primary (WTS, EXS, GDS) and so a child may be making totally normal and expected progress but wobble around grades if they are on the borderline.

Of course, the most extreme version of this is that you have only two grades: pass and fail. That’s how driving tests work, and we expect basically everyone to pass that sooner or later. This creates the ultimate cliff edge. Depending on what you accept, a fail becomes a truly damning grade, and likely for a bigger number of children who currently achieve a U. This feels pretty unfair for a kid who only just slips into that fail. More grades means that you can differentiate more at the bottom, as well as at the top.

A pass/fail would also have the consequence of making it almost impossible to motivate a very large number of children. If the expectation was that 90 or 95% of kids pass (after all, low pass rates is the problem it’s trying to solve) then how do you motivate the very large numbers of kids who know from entering year 7 that they are going to leave with their pass, and that no amount of additional effort will result in anything different. And again, qualification that is almost universally obtained looks and sounds a lot like a participation certificate to many outsiders, and it would be regarded with commensurate respect and esteem.

All of that notwithstanding, I have perhaps the most sympathy with a variant of this approach, which I’ll explore in a future post. In short, though, I’m quite a big fan of assessment approaches which are designed to have a very high pass mark in a relatively limited domain, and then to hold schools much more accountable for that pass mark being close to 100%. The phonics screening check is probably the best example of this. It is much harder to do as kids get older for lots of reasons, but I think that there are ways of mitigating this in a way that would be acceptable.

3. Scrap GCSEs or grades altogether.

This argument has gathered force since the law was changed in 2015 to mean that all children have to stay in education or training until they are 18. Given 16 is no longer the endpoint, many have asked why we have retained a high stakes terminal examination programme at this point. Couldn’t we just scrap GCSEs altogether?

Sometimes, this solution is presented in it’s purest form, not only would you abolish GCSEs, exams and grades, but you would scrap any sort of assessment at all. Kids would instead engage with lessons and learning for the love of it, unburdened by being probed and prodded, and instead able to focus entirely on the process of learning. Qualifications can come later, at 18, when they are required for the next steps.

I try very hard to be open minded but I think that this idea is bananas. Basically all of the bad things from above apply. It does not survive contact with the real world and anyone telling you that it does almost certainly hasn’t spent a meaningful amount of time in a school or with teenagers.

Usually, though, what people advocate for here is some new sort of assessment regime, which would be more suitably nuanced, positive and fair. It would recognise what kids could do, instead of putting them into neat little boxes. That kind of thing. Often some sort of e-portfolio is offered as an alternative, or perhaps some sort of coursework, or modular exams which accrue credit.

I won’t relitigate the ‘exams vs coursework’ debate here, but I believe it was easily gameable before genAI, and is now a complete non-starter on the grounds that there would be almost zero public confidence in it based on the fact that almost all children would be assumed to have cheated. E-portfolios would be a pretty good way of accurately assessing how much money kids parents have.

The least unfair

The consequence of all of these, by the way, would be felt the harshest by more socio-economically disadvantaged children. If confidence in GCSE qualifications collapses, then sixth forms, colleges, HEIs and employers will look for other signals that a kid may be worth taking a punt on. And most of those things are things that you can buy.3

This leads us to the truly unfortunate but unfortunately true conclusion that there is no perfect assessment system, but there are better and worse ways to assess kids depending on your objectives. If you want to go beyond lamenting a problem and instead propose a workable solution, you are very often simply making a selection of the least worse alternatives. Most of them will have been tried before, possibly quite recently.

I’ve looked at this hard, for what it’s worth. I really tried to talk myself into there being a magic different way, but have come to the conclusion that if fairness is an important criteria for you then, at the scale of a nation, exams really are the only game in town.

They are not completely fair. They are the least unfair.

4. Improve teaching and learning.

All of this leads us to the only other way of reducing the number of children who fail to demonstrate the standard for a grade 4: and that is to improve teaching and learning in schools.

This sounds awfully glib, perhaps even naive. You probably, like me, having a gnawing voice in the back of your head saying, “If we could just improve education then we would have done that. We’ve tried. Really really hard. This genuinely seems to be the best that we can do and it’s not moving the needle.”

But there are a reasons to be hopeful. Although it’s true that the number of kids not getting a grade 4 has remained relatively stable, the conditions that kids have lived under have not. Over the last few decades we’ve lived through the 2008 recession resulting in an austerity project that impacted not only school budgets but the state apparatus around families and in communities. Many, many families have faced difficult years and more challenging circumstances. Then the national curriculum and GCSE exams were reformed to make them much more demanding. In 2020, Covid meant two years of heavily disrupted schooling. And since then the wars in Ukraine and the middle east have continued to wreck the global economy an occupy almost all political bandwidth.

Given all of that, it’s kind of extraordinary that the pass mark hasn’t declined or collapsed. It’s like someone turning up the speed and incline on the treadmill simultaneously. The fact that you’re standing still doesn’t mean that you aren’t performing better.

More than this though, there are genuine improvements across subjects and grades. This feels quite counterintuitive because the numbers of children taking exams is very large. There were about 600,000 sixteen year olds taking GCSEs this year, which is a lot of people. All of us struggle to get our heads around very big or very small numbers, and statistics in these circumstances can be misleading or confusing (for example, if you double your rate of cancer by partaking in some vice, the original risk matters. Doubling from 0.003% to 0.006% doesn’t seem like a problem. Doubling from a 14% to 28% risk definitely does). This means that even quite small percentage differences in pass rates are in fact very significant. If the GCSE pass rate moved from 70.1% to 70.9%, that would be an additional 5,000 students getting a grade 4 instead of a grade 3.

The last reason is that we do now seem to have both the knowledge and infrastructure4 to make improvements at a scale which could shift the dial for meaningful number of kids. A pretty coherent set of institutions (such as the EEF, the Chartered College of Teaching, behaviour, attendance, early years stronger practice, English and maths hubs, the ITTECF and golden thread) exists to support schools and teachers in both understanding evidence on what is more likely to be effective, and in implementing change to to improve outcomes. The Free School policy allowed for a natural experiment of school models that result in more kids getting better grades. We have now a couple of dozen schools who have reliably secured passing grades for almost all of their pupils. And the Academy Trust structure means that schools are able to better support each other, and that support will come from leaders with expertise and receipts. Again, we are seeing some (not loads) of trusts reliably turning around schools that have historically struggled to get many kids a 4+.

We can do something

The emotional reaction to the annual third of kids is understandable. It would be convenient if it was statistically rigged, and abandoning norm-referencing would reveal that almost all kids are in fact at the standard expected for 16 year olds. But that’s not possible because it’s not how they are awarded.

So we have to make our choice of what we do do. There are some options which seem seductive at first glance, but would almost certainly actually make things worse for most or all kids in the long run.

Instead, we I think that we should back ourselves to continue our trajectory of improvement, look more closely at the schools who are outliers, continue to wisely innovate, embrace and deepen peer support systems and collaboration, and hold ourselves to account on getting all kids to an acceptable standard.

I’ve spent the last three years talking to people in education about belonging, and it has felt a bit like that at times. It’s clear that it’s a term that means a lot of different things to a lot of different people. There are books that are devoted entirely to different angles of it, some deeply personal, some academic, some focussed on adults, some on marginalised groups, the list goes on and on.

All of this means that if you’re already skeptical of a concept like belonging, it can confirm your worst assumptions. It’s a nebulous term, meaning everything to everyone and therefore basically meaningless.

I think that people are right to be skeptical, but wrong that belonging can’t be pinned down in useful ways, and that this isn’t really a priority in most schools today.

Muddying the water further is that there is now a fair old bit of commentary on belonging (often, let’s face it, from some grifters who have sniffed a school priority and so thrown together something in an afternoon that they might be able to pitch to some time-poor and anxious school leaders). And so, misconceptions and myths have arisen. Myths which could lead schools to spend time and money on practices in school that are unlikely to make any difference, and could make things worse.

Here are five of the most common myths that I’ve seen.

1. Belonging is about being ‘nice’ to kids.

It all sounds a bit happy clappy, and the head of RE who gets great results and is actually pretty well liked by the kids is now being encouraged to undergo a personality transplant to be all cuddly with the kids.

No.

Building the conditions for belonging is not the same as being ‘nice’ to kids. That’s not to say that being nice to kids is a bad thing, or should be avoided, but it’s just not really relevant here. Belonging is less about being nice and more about being unambiguous in the signals that we send to kids about whether or not they are welcomed and valued members of the group.

This relies in part on authentic interpersonal relationships and if kids can sniff a country mile off when a teacher is rehearsing a script or pretending to be someone that they are not. The point is made best by David Yeager, author of 10-25: the Science of Motivating Young People, who shared this when we interviewed him as part of the Building Belonging Project:

If you’re interested in more from Professor Yeager on Belonging, there are some highlights from his interview here.

2. Belonging takes time away from learning.

Schools have lots of different purposes but they are, and should be, places of learning first and foremost. It is what they are uniquely placed to do, and teachers and leaders are right to guard against the endless demands for their institutions to transform themselves into some other organisation with goals that aren’t passing on our inherited culture for the personal benefit of kids and collective gain of their society.

But belonging isn’t in tension with learning: it enables it. More strongly, the harder you want to push your kids academically, the greater the sense of belonging needs to be. That’s because genuinely challenging material means kids will struggle, and this could lead them to question whether they really belong in your class, which takes up cognitive resources.

So time invested in creating the conditions for belonging, which can often just be a tweak in language, is necessary for all kids to learn loads of hard stuff.

3. Belonging is just for some groups of kids.

This misconception is basically that belonging is just DEI or SEL in disguise. That immediately politicises it, but it also brings back memories for many of some pretty ropey practices which tried to address some genuine problems and maybe made it worse for the very kids it was trying to help.

To be clear, the lived experience of some groups of children will mean that they are more sensitive to belonging uncertainty in more situations. But one of the clear findings from research on belonging is that it’s a universal trait, likely rooted in evolutionary biology. All of us are sensitive to the question of ‘Is this a place where people want me around?’ because that question would have been existential for most of human history.

The work of Claude Steele and colleagues, explored in a previous post, shows that this uncertainty affects performance because scanning for cues to answer that question uses cognitive resources. Although they started with the experience of young Black men, researchers have since repeated the findings with women in the context of STEM, and white boys do worse in maths tests if they share the room with Asian children.

The safest assumption, therefore, is to assume that all children are unsure about whether or not they belong, and be much more explicit in terms of signals that they all are.

Like phonics, this harms none, benefits all and is essential for some. Evidence suggests that the effect will be disproportionately strong for kids that need it the most. And you remove the risk of emphasising particular in/out groups which could paradoxically worsen a sense of belonging uncertainty.

4. Belonging is the job of the pastoral team

One of the most common errors with addressing the needs of pupils with SEND is to make it one person’s job. In the worst cases, that meant teachers spending very little time with the children who required the most support, as they allocated them to teaching assistants to work with in a small group away from the rest of the class. The best schools have come to understand SEND as everybody’s responsibility. More than this, teachers spend the most time with these children. But everyone coordinates, SENDCO, AHT, HOD, teacher, support staff. All singing from the same hymn sheet, all supporting in a coherent way.

Belonging needs to be addressed in the same way. It can’t be something that gets added to a pastoral teams list of responsibilities, with the hope that a check in during form will fix the experience of that kids for the rest of the day. And that’s because belonging is ultimately predicated on interpersonal relationships. The first is the relationship between the teacher (and any support staff) and each pupil, and the second is the relationships between pupils themselves. If a kid feels known and valued by the adults in the room, they are more likely to feel as though they belong. If they don’t, they won’t, no matter how sensational the deputy head for culture is.

5. The school is the unit of belonging

Ok so this one is a bit more tentative than the others. Belonging has, in lots of schools, been identified as a whole school priority, perhaps featuring on the SDP, usually with the hope that it will improve attendance. That means that senior leaders try and answer the question of ‘do kids feel like the belong in this school?’ They then think about uniform, and assemblies, and values, and maybe extra-curriculars. And that stuff is important, but I think that its the periphery and not the core.

That’s because kids experience school at the classroom level. In primary, that will likely be a relatively stable experience: same room, same teacher, same group of classmates. In some senses that makes things easier (you simply. have more time to get to know kids and build strong interpersonal relationships, and establish rituals and shared norms that contribute to belonging). But it also presents a risk. If you don’t feel like you belong, or you feel a bit anonymous, then that’s your reality all day. When you ask primary colleagues to think honestly about their class, they can normally identify at least a couple of kids who fall into this invisible category.

In secondary, kids obviously move from classroom to classroom, and it’s perfectly possible, likely even, that their sense of belonging shifts pretty dramatically from lesson to lesson. This is not something that can really be fixed with a whole school initiative.

“Right, how are we doing belonging?”

I’d imagine that this question has featured in hundreds of SLT meetings over the last year or so. It’s a perfectly reasonable question to ask. Whether it’s in response to lagging attendance, fear of Ofsted, or a genuine desire to make the school a more welcoming and vibrant place, schools want to know how they can do belonging.

The problem is you can’t.

That’s because the question rests on a faulty assumption; that belonging is something that is external, to be delivered, or achieved. This misconception tends to lead to tokenistic and superficial fixes, like commissioning a mural saying ‘You belong here’, holding an affirming assembly at the start or the year, or having a drop down PSHE day.

These aren’t necessarily bad things to do, and are certainly defensible.1 It’s understandable. Schools are busy places. Leaders have a ton of competing priorities, and everything is important. And with the recent wave of interest in belonging, there’s no shortage of consultants or publishers willing to throw together a quick fix and sell you a dream of solving a legitimate and thorny challenge with a really quite pretty AI generated poster.

Belonging is a process

This is frustrating. It would be really nice to believe that belonging is something that is fixed, binary state, like whether or not a kid knows their 3 times tables or can tell you the four causes of the First World War. But belonging isn’t something that’s schools do, it’s something that children feel.

And not just children. All of us, all of the time, are either consciously or unconsciously making an evaluation of whether or not we fit in a particular environment. It’s part of being a human being, and probably traces back through our evolutionary psychology, where membership of the tribe would be life or death.

That means that we’re all constantly asking the question “Do I fit here?”, “Am I welcome?”, “Am I valued?” Gregory Walton coined the term ‘belonging uncertainty’ for this universal feeling. It’s not a charming quirk of teenagers, or a deficiency in small children. You probably experienced it when you started a new job, and maybe still feel it when you walk into the staffroom. I do it when I pick up my daughter from school and approach the group of parents at the gate.

It’s something we all do, and are particularly sensitive to in new environments, or environments with people who we don’t know closely.

The feeling of belonging, then, is an internal judgement. And we make that judgment based on cues from our environment. These cues could be physical, like whether or not there is a coat peg with our name on it, but mostly they are interpersonal; they are based on the words and actions of others.

Ambiguous signals

One of the big problems with making these inferences about whether or not we belong is that those cues are usually ambiguous. The teacher looks at you with a frown when you arrive? Could be they’re having a bad day, or maybe something you did last lesson annoyed them and they would rather you weren’t there. Friends start giggling in the corner as you sit down, maybe it has nothing to do with you. But maybe…

There are obviously thousands and thousands of these signals being received almost every moment. But they aren’t only ambiguous, they are also recursive. That means that once you begin to infer that maybe you don’t belong, you become more primed for signals that confirm that judgement. The same is true in the other direction, but for most of us there seems to be an asymmetry: we’re much more sensitive to negative signals.

Belonging isn’t a state, then. It’s a dynamic and continuous evaluation playing out in everyone’s head based on ambiguous signals from the environment we find ourselves in and the people that we share it with.

Expensive, simple, and wrong.

There are a couple of things that follow from this. First, the unit of belonging is the classroom. It may appear as a development priority on an SLT agenda, but it’s not something that we can meaningfully shift at a whole school level. Sorry.

Second, it means that belonging isn’t something that teachers do. It’s everything that they do.

This makes it different from so many other challenges, which can often be addressed with an intervention, or a better resource, or an additional member of staff. In some ways that’s pretty uncomfortable: there isn’t a programme that you can buy or a keynote speaker that you can hire that means you’ve ‘done’ belonging. As H.L. Mencken put it, for every complex problem there is an answer that is clear, simple and wrong. I’m not sure it gets more complex than human belonging, in classrooms, with large groups of children.

But in other ways it liberating. First, it’s cheap!2 It depends on how teachers interact with pupils and set up the conditions for them to interact with each other. That doesn’t cost anything. It’s also work that most teachers that I’ve met genuinely welcome. They got into the job because they like kids. They want them to succeed and make progress in the curriculum, yes, but they also want to get to know them, for them to look forward to coming into their classroom, and for them to feel safe, welcome and valued while they are there.

So no, you can’t do belonging. It lives inside children’s heads, based on some stuff that is out of your control, and plenty of stuff that the teacher can orchestrate. But the only way to minimise the chances of belonging uncertainty becoming the default judgment of most kids is through the culmination of a million totally ordinary interactions.

And if you’re thinking ‘well that all sounds lovely, but we don’t really have time for that - there’s stuff to learn!’ I hear you. But there’s quite a lot of evidence that a secure sense of belonging is a key enabler for pupils to tackle rigorous content.

In the next post, I’ll share a bit more about just that.

Subscribe

1. There is some evidence to suggest that these whole school initiatives could backfire though: if you’re uncertain about whether or not you belong, these heavy handed messages could confirm to you that you don’t. And none of them really get to the true root of the matter.

2. Ironically, this can be a problem. Schools tend to take more seriously, and invest more time and and attention, in things that they’ve paid for.

Share

Perhaps the seminal study that acts as the foundation of this work comes from Claude Steele, emeritus Professor at Stanford University. In his 2010 book “Whistling Vivaldi”, Steele shares an anecdote from author and Pulitzer Prize winning journalist Brent Staples.

Because as well as being a generationally talented writer, Staples is a Black man, and a Black man who in the 70s was, in his own words “a broad six feet two inches with a beard and billowing hair”. And so, when he walked through Hyde Park in Chicago at the night, on his way home from graduate school, Staples noticed something: White people would tense up, clutch their bags, or cross the street. Couples would squeeze each other a little tighter as he passed. So he started experimenting. When approaching other park goers, he’d whistle Vivaldi’s Four Seasons, or sometimes Beatles songs, as he walked. And what did he see? People’s reactions visibly changed. People’s shoulders relaxed, the air got thinner, and they stopped veering off to avoid him.

Staples was, he later realised, signalling something specific to passersby. He was telling them “I’m not a threat.” He did this by demonstrating that he was educated, which he bet would lead to the inference that he was therefore safe. (You can read Staples original paper from 1986 recounting the experience here).

Subscribe

Of course, it’s depressing to read this; that he needed to do it at all. The cognitive and emotional cost of constantly managing other people’s perceptions of you is enormous and ultimately, here, rooted in internalised racist assumptions.

Steele read about this and had a very specific question: how did this process affect performance. This took a human experience and turned it into a research programme, and one that is of huge relevance to educators. So, in a series of studies starting in the early 1990s, Steele and his collaborators showed that simply reminding people of a negative stereotype about their group, just before a test, was enough to harm their performance.

The first famous study, with Joshua Aronson began in the early 90s, and was published in 1995 under the title ‘Stereotype threat and the intellectual test performance of African Americans’. They gave Black and white Stanford students a difficult verbal test from the GRE. Half were told it was a measure of intellectual ability. The other half were told it was a problem-solving exercise that wasn’t diagnostic of ability. The white students performed about the same in both conditions. The Black students in the “ability” condition performed significantly worse than those in the “problem-solving” condition.

The specific stereotype here, of course, is the mendacious trope that Black people are less intelligent than Whites, which was a long embedded trope in the American story, and was seeing a resurgence with the rise of scientific racism in the 90s (the Bell Curve was published the year before Steele and Aronson’s study).

Spencer was interested in whether the impact of stereotype threat went beyond race, though. And so in 1999 conducted an experiment with women around maths, which was (and still is) a male dominated field. They found the same pattern. In the a study they preceded one maths test with some information about how men typically do better on the test, and they saw women’s performance drops at statistically significant levels. But when they told women that the test shows no gender differences, the gap disappeared.

The replication record on stereotype threat has been mixed in recent years, and some of the original effect sizes look smaller under closer scrutiny. But the broader insight has held up. There’s something about uncertainty around belonging, the wondering whether you’re really welcome and whether you’ll confirm what people already think, that makes it harder to do cognitively demanding work.

This is where the line runs from Steele’s stereotype threat research to the contemporary belonging literature. Greg Walton and Geoffrey Cohen took the insight and asked what schools could actually do about it. Their research showed that small, well-timed interventions, telling students that everyone struggles initially and that doubts about belonging are normal and pass, can produce sustained effects on academic performance, particularly for students from underrepresented groups.

For teachers and leaders, this points somewhere quite specific. It’s about understanding what’s happening in a child’s head when they walk into a classroomC and specially around cognitive load theory. We all now are well rehearsed in the idea that the more we ask working memory to process at once, the more likely it is that we will overload the finite resource, leading to confusion and/or task abandonment.

But when we think about cognitive load, it’s usually just in terms of the content itself: are too many steps in a procedure? Is there too much extraneous information on a slide? Is there novel vocabulary hitting kids faster than they can absorb it.

But there’s another kind of cognitive load that often goes unnoticed. If a child is wondering whether they belong in this classroom, whether the teacher likes them, whether they’ll be laughed at if they get the answer wrong, whether their peers think they’re stupid, that wondering is using working memory. It’s essentially a tax on cognitive resources, restricting the already limited bandwidth they don’t have for the lesson.

So although Belonging has a reputation for some as a fluffy moral goal outside of the business of learning, we can see how the origins of the research were grounded almost entirely in performance and the impact on cognitive process.

securing a greater sense of belonging is, I believe, an end in its own right (and I’m happy for it to be a fluffy moral goal). But if you only care about learning the most amount possible in the six hours that they are in school, then securing belonging buys you increased working memory available for the content of the lesson.

This reframes what we’re doing when we create classrooms where every child feels they belong. We’re making it cognitively possible for kids to learn what we want them to learn.

Brent Staples didn’t whistle Vivaldi because he wanted to entertain. He whistled it because the alternative was carrying exhausting levels of vigilance through every walk home, and deploying counter measures aimed at neutralising stereotypes. The irony, or tragedy, of what Steele showed is that doing this has an impact on performance which only acts as confirmation of the stereotype.

Maybe at least some of our pupils are doing something similar. They’re scanning, assessing perceptions whistling Vivaldi in classrooms, every day. And if that has an impact on their performance, they may just conclude that learning isn’t for them.

So the work of teaching is, at least in part, the work of making that whistling unnecessary.

Share

But why has ‘belonging’ become this season’s hot new trend? Like most fashions, the reasons that some ideas in education catch fire can be difficult to trace.

There are usually a couple of key impetuses, before companies and consultants sniff an opportunity, and then the fire begins to feed itself. Most headteachers or trust leaders don’t want to be the first to close their school when it snows, nor the last. So in a sense, everyone’s talking about belonging because everyone’s talking about belonging.

And I’m one of those people. Full disclosure, I have just finished a book on the subject. My decision to write it surprised some people. I’ve spent most of the last decade banging on about rigour and curriculum and ED Hirsch. In 2018, I wrote an article for Teach Primary humbly entitled “I’m Bringing Knowledge Back”.

So I was as surprised as everyone else when, in 2022, I decided to focus on belonging as part of an international fellowship with New America focused on the Learning Sciences. My group had 18 months to focus on pretty much any aspect of learning, and after some ‘ideation’ (*shudders*, *showers*) sessions we arrived at the conclusion that the biggest barrier, currently, to children making progress in their learning was a lack of belonging in school and classrooms.

(I should say here, that wouldn’t have been my conclusion 10 years ago, or even 5 years ago. In many ways, our attention can shift to belonging because the cognitive science and curriculum revolutions have become so well embedded into the educational architecture).

We spent those months speaking to experts from around the world - teachers, school leaders, researchers, parents, pupils and educationalists - and shared what they told us in this short video:

It’s honestly bittersweet that, since then, belonging has gained such traction. Sweet, because I genuinely think that it is something that can both improve children’s experience of school and contribute to progress in their learning. Bitter, because as soon as a new ‘thing’ becomes too popular in schools, it quickly gets hollowed out, twisted, and generally misunderstood and misused.

But back to the question of why? From where I’m standing, there are three main reasons that schools started to add belonging to their school development plans.

1. The Attendance Crisis

It’s getting better, but we still have way too many kids persistently absent from school. If a kid isn’t in school, then it doesn’t matter how well the curriculum is sequenced or how slick instruction is. Those kids are learning nada.

We should be clear here: the attendance crisis is not an England specific phenomena. It’s happening around the world. And it’s not entirely a ‘covid problem’. Attendance was dropping and persistent absence was rising even before the pandemic.

So some school leaders have arrived at ‘belonging’ as at a way to try to address this specific problem. If kids aren’t attending school, it must be because they don’t feel like they really belong. They must not feel safe, known, recognised or valuable. Because if they did - they’d come.

2. Ofsted

Ofsted put it in the framework. Yep.

3. “This isn’t the school I want to lead”

Some leaders have spent hard years optimising for maximum efficiency in terms of learning. They have codified anything that could possibly be codified. Teachers rehearse and script and align. Kids are drilled into slick routines. The curriculum is specified at to a forensic level of granularity.

And that has resulted in some astonishing successes. Kids who historically would leave school without formal qualifications are leaving with a massive menu of options and opportunities. I’ve been lucky enough to spend some time at schools overseas, and the quality of curriculum thinking and clarity around effective teaching in England is genuinely point of national pride.

But. Here’s the but. I’ve spoken to quite a few school leaders who walk around their school and feel that the place doesn’t really have any soul. They think that their kids learn loads, but don’t think that the place has any real character, that kids can really be themselves. This doesn’t mean that they want to throw out the retrieval practice and turn the place into a youth club, but they do feel as though greater attention is needed to what it means to ‘belong’ to the school - what are you ‘belonging’ to?

In some ways, the reasons for belonging rising up the agenda for school leaders are not that important. It’s having its moment, and it will run its course. What I want to focus on is how that happens. I feel a huge sense of gratitude that England had its ‘curriculum revolution’ over the last decade, as well as embracing cognitive science and its applications to the classroom. But the key reasons for the success of these projects were:

a) they were grounded in evidence, with a theory base widely discussed, debated and disseminated

b) they were taken seriously by serious people, who took time to work out what was being claimed, and what could be claimed

c) they didn’t promise to solve everything in schools, but rather focused on one area that had become neglected

d) they worked within the realities of actual classrooms, and all of the constraints that schools have, builing practical, replicable and scalable applications.

That’s what I’d like to do with belonging. The Building Belonging Project was our first step towards that, and my book will be the second. Over the next few months I’ll be using this substack to share some of the research, ideas, case studies and progress around belonging. If that’s the sort of thing conversation that you’re up for joining, then please do subscribe.

Subscribe now