Does assessment fail the test?

Are we prepared to stop pretending we can weigh the pig into plumpness?

This week, we take a brisk stroll through the minefield of school assessment in the UK. Politicians are murmuring about reform, the Guardian is foaming at the mouth, and students… well, they’re apparently either panicked or bored. So, where are we really?

---

Assessment, like death taxes and school shoes, feels inevitable. No one seems especially fond of it, but everyone insists we need it. It is, we’re told, the cornerstone of rigour, the proof of learning, the means by which we separate wheat from chaff. Yet as the DfE embarks on yet another review, as newspapers trot out the latest tales of GCSE-induced trauma, and as university professors fret about the moral collapse heralded by ChatGPT, we might pause to ask: what exactly are we doing when we assess?

The Great Review

The Department for Education’s long-awaited curriculum and assessment review, chaired by Becky Francis, has released its interim findings. (More on David’s thoughts here.) The document is couched in the usual technocratic language - “misaligned systems,” “future readiness,” “equity of progression” - but between the lines there is a damning admission: the current assessment regime is not fit for purpose.

Key concerns? The exclusion of SEND pupils, the shallowness of curriculum coverage, the irrelevance of much tested knowledge, and the sheer inequity of post-16 outcomes. Stakeholders, from the Royal Society to the National Literacy Trust, echo the refrain: assessments must do more than sort and shame. They should deepen learning, support literacy, and widen opportunity.

What’s telling, however, is how cautious even this ‘review’ is. The critique is clear, but the proposals remain hedged, gesturing toward incremental reform rather than the radical reimagining the moment seems to demand.

A ritual in search of a purpose

the purpose of exams is not simply to measure learning but to provide an objective, external check on what students know and can do. Exams create a shared standard across schools, offering comparability, consistency, and transparency in a way that internal assessments cannot. They help validate effort, structure curriculum, and provide motivation for students.

But, when viewed through a behavioural lens, the apparent logic of exams is simple and seductive: what gets measured gets done. Students revise because there is a test. Teachers teach to the specification because that’s what will be assessed and how they will be held to account. Schools prioritise subjects that count in league tables. In theory, the exam acts as a motivator, an accountability tool, and a behaviour management system all at once. The assumption is that without this pressure, effort would dissipate and standards would slip.

Exams exist to sort and rank students. The logic rests on the belief that education must discriminate between students. Exams, under this view, serve to stratify: they separate the high achievers from the rest, allocate future opportunities, and create a hierarchy of merit. This sorting function is seen as necessary for managing scarce resources - university places, job offers, sixth-form admissions - and for signalling ability to institutions and employers. In essence, exams don’t just measure learning, they ration opportunity. The risk, of course, is that we begin to value the ranking more than the learning itself.

What are the effects?

The consequence appears to be a dramatic decline in students’ mental health. According to one survey, 85% of UK students experience exam anxiety, with 15% showing signs of high test anxiety and 30–40% of pupils report high anxiety levels during exam season. A synthesis of 48 studies concluded that exam periods are consistently associated with heightened anxiety and poorer mental health . Moreover, Childline reports a 10% increase in counselling sessions specifically focused on exam stress since the pandemic, totalling nearly 2,000 sessions in one year alone.

More recently, a 20024 survey by the Association of School and College Leaders (ASCL) paints a stark picture of the emotional toll GCSEs are taking on pupils. Drawing on responses from 787 school leaders, the findings reveal that 73% report students are “significantly anxious” about exams, with nearly half (48%) witnessing pupils leaving exam halls in distress. A further 45% said pupils required special arrangements—such as extra time or separate rooms—due to the severity of their anxiety. ASCL described the situation as a “mental health emergency playing out in school exam halls,” calling into question the cost of the current high-stakes model.

So what should we do?

The question of whether GCSEs should be scrapped has never felt more urgent or more divisive. A recent Guardian feature gathers a chorus of voices from across the education landscape, revealing deep disagreement not only about the exams themselves but about the broader purpose of education.

For those in favour of abolition, the argument is clear. GCSEs are seen as outdated qualifications that reflect an education system designed for a time when most students left school at sixteen. That time has passed, but the exams remain. Headteacher Ben Davis describes GCSEs as actively harmful. He says, “I’d go so far as to say they are making young people ill and they impoverish teaching.” His alternative is a diploma model that blends academic subjects with vocational routes and life skills, allowing pupils to build a portfolio suited to their strengths and aspirations.

Columnist Simon Jenkins goes further, describing the current system as a destructive cult that ranks and controls rather than educates. He urges policymakers to face the truth and abolish the qualification altogether. Lee Elliot Major, professor of social mobility at the University of Exeter, Says, “We’ve created an exam system that perpetuates privilege from one generation to the next.” In his view, GCSEs do not reward merit, they reward resources. Private tutoring and middle-class cultural familiarity tip the scales, while those without access to these advantages are left behind. This sentiment is shared by head teachers of various private schools, and, of all things, the Tony Blair Institute. Their message is the same: the system is no longer fit for purpose.

Others, however, are more cautious. Sammy Wright, author of Exam Nation, argues that GCSEs push all students into an academic mould and only offer alternatives once those students have failed. He proposes a “passport qualification” that would reflect a broader set of outcomes and provide a more inclusive picture of achievement. Mary Richardson at UCL’s Institute of Education echoes the need for reform. She sees the issue less in the assessments themselves and more in the culture that surrounds them. GCSEs, she suggests, have become part of a corporate machine that adds little value to teenagers’ lives. Her solution is greater investment in teacher assessment and more meaningful forms of evaluation.

Not everyone is convinced. Katharine Birbalsingh, head mistress of Michaela Community School, defends GCSEs as an essential part of a rigorous education. Exams, she argues, do not just measure students. They keep schools accountable. Without them, she fears, standards would collapse into a kind of educational free-for-all. Richard Brown, headteacher at Urswick School, takes a similar view. He admits the system is not perfect, but insists it offers pupils a real and motivating challenge. He describes GCSEs as a “powerful brand,” and believes many students thrive because of them.

The tension between these views is not easily resolved. On one side is the call for a more humane, flexible, and inclusive system. On the other is a belief in clarity, comparability, and structure. While the debate gathers pace, the reality remains unchanged. Students continue to line up outside exam halls. Teachers continue to hand out pens. The questions may change, but for now, GCSEs lumber on.

Where the research points

Christopher DeLuca’s article, Assessment in education: looking back, looking forward marks thirty years of the journal Assessment in Education, offering both a retrospective and a provocation. DeLuca traces the shift in assessment culture from broad, formative aims to narrow, high-stakes accountability. The trend, he argues, has been towards ever more standardisation, driven less by pedagogical insight and more by political anxiety. The result has been a system that constrains teaching, narrows curricula, and reduces learning to a performance.

The article laments the dominance of what it calls “assessment for surveillance.” Rather than supporting learning, assessments have become tools for managing teacher behaviour and policing school performance. This shift has hollowed out teacher agency and turned students into data points.

Yet the tone is not purely elegiac. DeLuca also calls for a return to purpose. Assessment, he insists, should be reimagined as a process of dialogue, not judgement. It should be embedded in classroom practice, responsive to context, and centred on learning rather than sorting. The future may lie in rebalancing professional trust with public accountability, and in resisting the illusion that numbers alone can capture what matters in education.

The problem is, embedding the assessment of students’ time in school in classroom practice and professional judgement comes with some predictable costs. Exams, despite Lee Elliot Major’s claim that GCSEs “fail the fairness test,” remain the fairest of all the unfair options available. As I’ve argued elsewhere, exams are standardised, externally marked, and less vulnerable to unconscious bias, social influence, or parental pressure. Teacher assessment, by contrast, is shaped by relationships, expectations, and sometimes by a desire to be kind, all of which can end up entrenching inequality rather than addressing it. Paradoxically, the impersonality of exams is what gives them their relative fairness: everyone sits the same paper, under the same conditions, and no one can be marked up (or down) because of who they are.

There’s strong empirical support showing that standardised exams produce fairer outcomes, especially for disadvantaged pupils, compared with teacher assessments. A 2021 Ofqual literature review concluded that teacher-based assessments are “more vulnerable to bias” than test-based assessments. It found consistent evidence that teachers tend to grade less advantaged students lower, even when those students perform equally well on objective tests  . Similar findings emerged internationally: students from low socioeconomic backgrounds or with special educational needs often receive lower teacher-assessed grades compared to their test-based performance. Daisy Christodoulou has summarised the historical case in plain terms: standardised exams consistently benefit disadvantaged students, challenging the common assumption that only teacher assessment helps those from underrepresented backgrounds. In fact, she argued, exams have historically been the mechanism through which civil service and educational systems sought to limit patronage and privilege .

So, Where Now?

We find ourselves in a strange position. Everyone - students, teachers, researchers, even some politicians - agrees the system is creaking but we remain paralysed by habit, tradition, and the deep political appeal of a number in a spreadsheet.

We speak of rigour but reward obedience. We value data over depth. And we persist, stubbornly, in assessing children not as they are, but as we once imagined them to be. The question isn’t whether assessment needs to change, but how. If exams, for all their flaws, are the fairest form of assessment, we have to focus on creating tests that are as good as possible and suit all children, not just the academically able. The idea that 40% of students fail English and maths every year cannot be acceptable and does npt have to continue. The are better, fairer ways to design and use exams.