The Covid-19 pandemic has brought the system of school examinations into sharp focus.1 The examination season for A levels and GCSEs was cancelled over two consecutive summers in 2020 and 2021.2 In both years, examinations were replaced by teacher-assessed grades—but only after a fight against the attempt by Boris Johnson’s Conservative Party government to determine students’ grades by algorithm. In the summer of 2020, students demonstrated against the unfairness of using algorithms to assign grades and won a very significant victory when the government backed down. One year later, in the summer of 2021, exams were once again cancelled and replaced by teacher-assessed grades, though this time the government made no attempt to use an algorithm due to its embarrassing retreat the previous year.
The pandemic opened a space for assessment in schools to be done differently. However, it also exposed (to those willing to look) the underlying unfairness of the existing school examination system. This article will explore the history of school examinations, the role they play in society and why they have become such a dominant element of the education system.3 It will also probe the ideology of ranking and referencing behind the failed algorithm. Finally, it will ask whether school education would be better off without a system of examinations and grades.
Why have exams?
Most societies and social systems require assessments of some type. Indeed, few would argue with many of the stated educational objectives of examinations: consolidating learning, maintaining a consistent standard of performance, providing feedback to the teacher and assessing the intellectual ability of students. Yet, this raises the question of how much examinations really fulfil these objectives. Moreover, if they fail to do so, then why do exams remain so central to the education system?
Examinations in whatever form they come have certain functions in common. Public examinations serve as an extrinsic source of motivation by attributing different levels of success and failure to candidates. However, the grading system itself can and does have a prejudicial effect on the students under consideration and distorts information about them. The summation of a student’s varied abilities into one exam grade or mark is bound to entail the loss of valuable information about the student, undermining the stated aim of providing feedback to the teacher. This feedback is simply replaced by a grade, and this grade is felt to sum up the student’s qualities. Unfortunately, we are all prone to the idea that a quantifiable grade can actually tell us more than it does. As A level student Aliyah York from PupilPower, a school students’ advocacy organisation, explained on Radio 4’s Rethink Education programme:
People look at your grades, and it determines your worth. You know nothing about me as an individual. You just see a set of numbers on a piece of paper and that’s it… I remember being so gutted on GCSE results day when a couple of grades did not reflect what I was capable of… I could not go onto the school I wanted because I’m not seen as worth it or as valuable—all based on just one exam. That is just wrong.4
“Success” thus tends to be a measure of the ability of the student to perform in examinations rather than their general intellectual ability. Moreover, there is also the widely evidenced phenomenon of self-fulfilling prophecies, when a teacher’s expectations, often based on past exam grades, determine the future success of a student to an alarming degree.
The grades a student achieves are reported as information to parents and teachers, Ofsted inspectors and employers. Yet, even here there are apparently contradictory values at work. For instance, employers frequently complain that young workers do not have the maths and English skills they require despite having good grades.5 Whatever the grading function of exams for employers is, it cannot simply be a means of checking that the student has grasped a sufficient number of facts and skills to make them fit for the job.
Grades and the inevitable attendant stress on success and failure can have no basis in seeking the consolidation of learning and accessing the intellectual ability of a student. Indeed, grading by competition is the exact antithesis of measuring the ability of a student against a supposed rational or “objective” set of criteria. The exam becomes an end in itself. The consolidation of learning becomes replaced with short-term memorisation of knowledge and exam technique. So, grading appears to reinforce aspects of assessment that run contrary to its stated aims. Yet, the competitive aspect and the possibility of failure are absolutely central to the exam system. On the aforementioned episode of Rethink Education, former Independent editor Amol Rajan argued, “Exams…are fantastic for character development. There is a moral value to failing in exams. They connect effort with reward…and they send a signal to employers.”
Grades achieved in public exams at 16 and 18 stay with people for life and have profound effects on a young person’s sense of self-worth. The sense of failure after not achieving the required grades can have a powerful impact on mental health and is perhaps the single most destructive aspect of the entire examination system.
Nevertheless, it is also clear that grades are not required if all you want to do is consolidate learning and access the intellectual ability of a child. This does not lead to a rejection of all examinations and assessment. Contributing to a rank and file teachers’ pamphlet in 1979, Richard Noss wrote:
No one but a fool would allow an unqualified surgeon to wield a knife…or allow untrained scientists to experiment with potentially dangerous chemicals… Insofar as such assessment is carried out by means of some kind of examination, few would argue with the idea.6
The idea of assessment itself is not the issue, but rather the nature of the examinations within our schools and the role that they play in the reproduction of the current social order.
Examinations and tests now dominate the school curriculum, acting as an inbuilt conservative force that limits innovation, diversity of ideas and learning for its own sake. Teachers will by and large teach to the examinations. Any teacher who decides to deviate in any significant way from the content of the examination syllabus would soon be in trouble with students, parents and headteachers. This domination of the education system by examinations needs to be explained.
Why have exams become so dominant?
In order to understand why examinations have become so all-pervasive, we need to look beyond the horizons of the education system to the society that shapes it. In any society it is necessary for certain skills to be acquired by certain people; the primary objective of a mass education system is to produce tomorrow’s workforce.7 The level of education required by this workforce changes over time, and that affects the demands on the education system. The levels of numeracy and literacy required by today’s society and economy are much higher than they were 100 years ago, and new demands for creative thinking and problem solving have emerged.
However, our society requires not just a workforce with the requisite level of education, but also an education system that reflects and actively reproduces key aspects of the prevailing social relations. Education takes place in a hierarchically organised capitalist society, based on competition and exploitation.8 Exploitation of one class by another class is hidden behind market relations but its effects are clear to see in the inequality that surrounds us. Exploitation means that future workers, who will exchange the use of their skills for a wage, must be socialised into an acceptance of their place. Examinations—with their emphasis on competition, grading and failure—are an effective and important way of inculcating such acceptance.
Although the reproduction of class inequality also takes place in the state school sector through setting and banding, it is most clearly seen in the divide between state and private education.9 About 7 percent of young people have a private education and 93 percent attend state school. Yet, the Social Mobility Commission’s “Elitist Britain” study found that a tiny privately educated elite—including many who went to the Universities of Oxford and Cambridge—continue to dominate top professions. The researchers looked at the educational backgrounds of around 5,000 leading figures including politicians, tech bosses, cultural stars, journalists, judges and chief executives. Around 39 percent of those in these jobs have had a private education.10
In politics, the study found that 39 percent of the cabinet went to fee-paying schools at the time the analysis was carried out. Some 65 percent of senior judges, 57 percent of members of the House of Lords, 59 percent of civil service permanent secretaries and 52 percent of diplomats come from a private school background. Around 43 percent of the 100 most influential news editors and broadcasters and 44 percent of newspaper columnists went to fee-paying schools; 33 percent of those went to both private school and either Oxford or Cambridge. In the arts, 44 percent of top actors and 30 percent of pop stars went to independent schools. Still, the figures were somewhat worse 40 years ago:
In 1984, at least 83 percent of High Court and Court of Appeal judges, 69 percent of ambassadors, and 89 percent of law lords had been to public schools. The financial elite came from the same sector. In 1981, 92 percent of directors of major life insurance companies, 70 percent of directors of clearing banks…70 percent of Tory MPs and 14 percent of Labour MPs came from private schools.11
Class inequality is also maintained through the role of schooling in transmitting ideas, beliefs and patterns of behaviour. These ideas and beliefs are communicated via the exam-dominated curriculum. A role is also played by what US economists Samuel Bowles and Herbert Gintis call the “hidden curriculum”.12 In Schooling in Capitalist America, they argue that there is a correspondence between the values learnt at school and the way the workplace operates.13 These values are taught through a “hidden curriculum” that children learn through the experience of attending school rather than the main curriculum subjects. This includes pupils unquestioningly obeying teachers, motivation by external rewards, pupils having no control over what they are taught and blame for failure being transferred from the education system to the shortcomings of individual pupils.
In this light the contradiction between the stated and actual effects of examinations does not seem so strange. We inhabit a society where the day-to-day work of the overwhelming majority of the population is what Marx called alienated labour, and the main motivation to work is the payment of a wage.14 The education system reflects these features of the capitalist social system. The very nature of what is to be learnt is itself often “alien” in terms of its cultural and ideological basis. Pierre Bourdieu used the term “cultural capital” to describe a type of schooling in which the curriculum is designed to favour those with privileged cultural backgrounds. Examinations ostensibly allow equal opportunities, but in reality they are geared to the cultural values of the ruling class, allowing this class to perpetuate their privileged position by giving their children a head start towards success.15 It is thus unsurprising that public exams need an external reward in order to motivate many students to work.
The dual aims of capitalist education are to provide tomorrow’s workforce with the level of education needed by capital and to ensure that children are socialised into knowing their place. These twin goals have meant exams playing an increasingly central role in schooling. The development of the examination system has been slow, piecemeal and often chaotic. However, the usefulness of examinations—both for selection and as a reinforcement of capitalism’s competitive, “winner takes all” values—has rarely been in doubt.
Examinations in the 19th century: payment by results
Examinations in one form or another go back many centuries. However, it is only with the demands capitalism made on educational provision in the 19th century that assessment procedures began to become responsive to the needs of the economy. Even then it would take over 100 years for a universal system of school examinations to develop.
In the early 19th century, the growing need to evidence competence in some professions led to a shift away from basing occupational roles and university places on hereditary wealth and breeding. In 1815, the first professional qualifying exams were instituted by the Society of Apothecaries in order to ensure that doctors were adequately trained. Exams for solicitors followed in 1835 and accountants in 1880. Sociologist Patricia Broadfoot comments:
The institution of examinations, related quite specifically to a particular vocation, marked the beginning of a trend away from taking ascribed occupational roles based on hereditary wealth and breeding alone—or lack of it—to a situation in which such roles were, at least ostensibly, the result of individual achievement and merit.16
These exams were a way of controlling entry to professions. In the growing number of 19th century elementary schools, where the mass of ordinary young people, destined for manual labour, were educated, there were no final exams or even leaving certificates. Pupils were drilled in the “three Rs” (reading, writing and arithmetic). Schools were held accountable by the so-called Revised Code for Education and administered through a system of “payment by results” in which they were given money in return for bringing individual pupils up to the required standard.17 The pedagogical result of the payment by results approach was teaching and learning by rote. The tests used “criteria referencing”, that is, measuring a pupil’s performance against pre-determined criteria and learning outcomes. The use of these criteria was confined to an exercise of quality control rather than selection. There were often written descriptions of what pupils should know and be able to do, and pass rates—as well as payments—could vary from year to year depending on pupil performance.
The growing number of grammar schools at the time accommodated aspirants for the newly created clerical, scientific and managerial jobs. Since success was limited to the number of scholarship places available rather than the number of pupils who could reach an agreed standard, scholarship examinations for entry to grammar school in the 1870s could not use criteria referencing. Instead, they operated a system of “norm referencing” whereby the percentage of children who would pass was decided in advance. This is the exact antithesis of measuring the ability of a student against some supposedly rational or “objective” criteria.
Examinations in the first half of the 20th century: school certificates and scholarship tests
The situation at the start of the 20th century was described by the Board of Education as a “state of chaos characterised by a range of examinations that was simply bewildering”.18 Researcher Val Brooks comments: “Free from state control, examinations sprang up on a makeshift basis. Some were academic and general in nature, but schools also used qualifying exams established by the professions”.19 Not everyone welcomed the increased use of examinations. At an early stage, the undesirable effects of exams on the curriculum were recognised, including by educationalist Edmond Holmes in his work What Is and What Might Be, published in 1911.
However, the need for standardisation and some measure of order led to the introduction of the School Certificate in 1917, which was taken at 16, and the Higher School Certificate, taken at 18. Both exams were reserved for grammar schools and overseen by universities. Elementary school children who left the education system at 14 and even fee-paying students who had been selected for a secondary education finished without any qualifications. However, as Broadfoot notes:
By 1922 and the establishment of the Higher School Certificate, a rationalised system of competition, norm-referenced school examinations was firmly established through which all but a very few professional and academic aspirants must pass.20
This put an end to school examinations linked to particular professions. The School Certificate was a standard qualification for those leaving school and seeking entry to university. This shift meant that the content of assessment came to have little to do with the nature of the activity for which it was acting as a selective procedure.
It is important to emphasise the very small numbers of pupils who were actually affected by the introduction of the School Certificate: “The number of pupils who emerged from school with a qualification School Certificate, was tiny—4 percent of the 16-17 age group”.21 The majority of pupils, destined to be manual workers, were denied even the opportunity to compete with their more academic peers for external certificates. Over 90 percent of young people did not receive a secondary education and consequently were not allowed to take examinations: “Prior to the Second World War, just short of ten percent of the population received a secondary education, and the majority completed their education in elementary schools, where the leaving age was just 14 years old”.22
The changing needs of capital and the growing number of clerical, scientific and managerial jobs led to a rise in the competitive scholarship tests.23 It was now necessary to select children who were thought to be “intelligent” and would benefit from an academic education. This raised the thorny question of how to identify academically able children and at what age this should be done. Brian Simon argues that, in the scholarship tests:
The relevant curriculum was taken as given and as a yardstick for diagnosing “intelligence”. The children who could “take it” were intelligent, and those who could not lacked ability.24
One of the main “administrative” problems to be solved was to find a way of fitting the selective procedure to the actual number of selective places available. There was a need to discriminate among pupils on educational grounds rather than on social ones. Broadfoot argues that it is “from these ideological and pragmatic pressures that the concept of ‘meritocracy’ was born”.25
The search for a means of implementing this supposed meritocracy involved finding some objective measurement of merit. The solution was ready to hand and found in the new “science” of mental testing that was establishing itself in the first half of the 20th century. A leading light in the emergence of intelligence testing was Cyril Burt, a psychologist working for the London County Council. “Intelligence”, Burt wrote in 1947, “will enter into everything the child says, thinks, does and attempts, both while at school and later on… If intelligence is innate, the child’s degree of intelligence is permanently limited.” He continued, “Capacity must obviously limit content. It is impossible for a pint jug to hold more than a pint of milk, and it is equally impossible for a child’s educational attainments to rise higher than his educational capacity permits”.26
If this was the case, then all that was required was some way of measuring this innate intelligence and Burt provided the solution with his “intelligence quotient” (IQ) tests, which later formed the basis of the “eleven-plus” test. There were few doubts or questions about all this in the early 1930s as the “science” of mental testing as a way of measuring innate intelligence gained ground. Not only was intelligence innate and unchanging according to Burt but it was also inherited. As Broadfoot points out, “Intelligence testing, as a mechanism of social control, was unsurpassed in teaching the doomed majority that their failure was the result of their own inbuilt inadequacy”.27 What marked Burt’s work out from other psychologists working in the field was the mass of data and the strength of the correlations he found. His ideas were so convincing that they held sway for over 30 years and underpinned the reforms to the education system introduced in 1944 after the publication of a state-commissioned report by educationalist Cyril Norwood. The three types of school distinguished by these reforms (grammar, technical and secondary modern) allegedly reflected the three types of mind among children:
The main set of arguments grounded the proposed restructuring of secondary education in the nature of the child. Some (a few) were capable of abstract thought and interested in ideas and “learning for its own sake”—for these grammar schools should be provided. Others (also a few) were more interested and adept at the application of ideas in technology—for these there should be (selective) technical schools. The great majority, however, were more concerned with practical activities and the immediate environment—for these the new type of secondary “modern” schools should be designed. The Norwood Committee claimed that they based this proposal on “child-centred” ideas. Each of the three types of school were needed to “match” the nature of the child.28
There is a clear correspondence between the supposed “nature” of the child and the demands that capitalism was making on the education system at the time. These stressed the need to educate a minority suitable for the professions, a growing minority of workers able to work in technical, scientific and clerical roles, and a majority that would continue to perform manual labour.
Exams in the second half of the 20th century: O levels, A levels and the eleven-plus test
The school leaving age was raised to 15 in 1947, and the eleven-plus exam—based on Burt’s IQ tests—introduced to select pupils for a grammar education. With the new structure of schools came the introduction of a new examination system in 1951, and GCE O levels and A levels replaced the School Certificates.29 These new subject-based certificates allowed greater flexibility in the curriculum. Candidates were originally awarded only a pass or fail classification at O level. The grading system for O level were introduced later and allocated on a norm-referenced basis. The new examinations allowed for finer differentiation and greater specialisation, better serving an economy with an increasing need for different kinds of expertise.
From the start there was a conscious policy of excluding working-class children from exams. This was based on the theory of children having specific “natures”, and the intention was to restrict access to the new exams to grammar school students. O levels had an age limit of 16, effectively excluding secondary modern children, who left school at 15. Ellen Wilkinson, Labour’s education minister under the Clement Atlee government in the late 1940s, actually set the GCE O level pass mark higher than the old School Certificate, because she wanted to prevent the new secondary modern schools from entering pupils for it.30
However, by the early 1950s, pressure from the secondary moderns for access to external examinations was growing. The central role of examinations in determining career opportunities made it impossible for secondary modern schools to remain uninvolved in the competition. Parents and pupils pushed for a chance to compete. At the same time, the status of individual secondary moderns became increasingly dependent on how well pupils did academically, and they tried to imitate high status grammar schools. O level and A level GCEs remained the only officially accredited routes to higher education and the professions.
The Certificate in Secondary Education
As an increasing number of children were being entered by secondary modern schools for GCE O level, an examination that was meant to be the preserve of the grammar schools, a new examination, the Certificate in Secondary Education (CSE), was introduced in 1965. The Marxist educational historian Brian Simons explained the official thinking of the time:
For the top 20 percent of the total 16 year old age group (that is, those in selective schools), the GCE O level provided an appropriate objective…the new exam proposed would be designed for the next 40 percent. This 40 percent was divided into the top 20 percent who could be expected to pass in four or more subjects and the next 20 percent who “might attempt individual subjects”. No examinations were proposed for the “bottom” 40 percent.31
This tripartite system of schooling proposed under the 1944 Education Act never really had time to establish itself before the emergence of a new form of school, the comprehensive school, which replaced the secondary modern and grammar school. At the 1963 Labour Party conference, Harold Wilson, soon to become Labour Prime Minister in 1964, argued:
To train the scientists that we are going to need, we will need a revolution in our attitude to education—not only in higher education but at every level… As a nation we cannot afford to force segregation on our children at the eleven-plus stage… We cannot afford to cut off three quarters of our children from virtually any chance of higher education.32
The need for greater numbers of young people to progress into higher education together with the demand for equality of opportunity for all young people resulted in the selective grammar schools gradually being replaced by non-selective comprehensive schools that took all children from their local catchment areas. The spread of comprehensive schools led to a phasing out of the eleven-plus test in most areas during the 1960s and 1970s. At the same time, Burt’s theories of intelligence were becoming increasingly widely questioned. As Broadfoot describes:
During the 1950s the concept of intelligence as fixed and inherited gradually became untenable as a result of a series of research studies. These showed, among other things, the social class and environmental influences on intelligence, demonstrating that the IQs of pupils in grammar schools improved while those of their counterparts in secondary modern schools deteriorated.33
Both intelligence tests and Burt himself were both severely discredited in the 1970s when investigators found he had invented collaborators and that his work had involved fraud on a massive scale.34
The use of the eleven-plus test to determine whether pupils went to grammar or secondary modern nevertheless continued in some local authorities. However, attempts were made to improve the number of girls at grammar school, where historically there had always been more places available for boys. A problem arose when girls started to do better than boys in the eleven-plus test. There was then a conscious policy in some local education authorities of actively marking girls down to ensure a balance of numbers between the sexes in grammar schools. According to educationalist and Labour Party politician Frances Morrell:
The test figures…were adjusted so that an equal number of boys and girls were represented in the top ability range. As a result, some girls were refused admission to grammar schools despite having test scores higher than some of the boys who were admitted.35
The girls were informed they had done less well in the test than they actually had. The idea that girls could not do better than boys was so ingrained that the data was simply changed to fit this prejudice.
Those fighting against discrimination in the examination system did not just have to battle against the school structure. They also had to contend with the way examination questions in some subjects reflected the dominant ideas in society. The teacher and women’s rights activist Audrey Jones recalled some of the most sexist examination questions:
The outstanding one was in home economics: “Your brother has been out for a football match and brought a friend home for a meal. Describe the steps you would take to arrange for their laundry and the meal you would provide for them”.36
The situation was even worse for black children. Culturally biased IQ tests were used as a basis to wrongly send large numbers of West Indian children to schools for the “educationally subnormal” (known as ESN schools). This state of affairs was powerfully described in Bernard Coard’s How the West Indian Child is Made Educationally Sub-normal in the British School System: The Scandal of the Black Child in Schools in Britain.37
In some ways these sexist and racist practices ran counter to the overriding need of capital at the time, which was to increase the supply of scientists and technicians in order to meet the demands of the economy. This motivated the raising of the school leaving age to 16 in 1974. This allowed all children the opportunity to take examinations since O levels and CSEs were restricted to children of 16 years of age.
By 1976, 75 percent of secondary pupils were in comprehensive schools.38 Nevertheless, as Simon makes clear, the comprehensive system preserved the divisions that had preceded it:
No more than 50 percent of the existing schools were “genuinely” comprehensive… Further, within both the “genuine” comprehensives and the others, internal divisions were still imposed through the continued existence of the dual examination system (GCE and CSE), which was originally devised for the divided structure existing before comprehensive reorganisation.39
All this notwithstanding, the threefold division of children into GCE students, CSE students and the rest opened up a space in which progressive teachers could experiment with different types of pedagogy and assessment. The new CSE had three different modes of assessment. One particularly favoured by progressive teachers was the Mode 3 CSE, which gave teachers and schools more control of assessment methods. Within a framework of continuous assessment there was scope for mixed-ability teaching, cooperative working and project work, as well as possibilities for self-expression and experimentation on the part of the students.
The result of all this was that an increasing number of school leavers left with qualifications: “By 1982, over 25 percent of all school leavers had five or more O level passes, while 20 percent were achieving an average C grade in seven or more subjects”.40 Given this, there was a growing strength of feeling among comprehensive teachers in support of the introduction of a single exam for all at 16. This would replace the divided system of GCE O levels and CSEs, which was only ever designed to involve the top 60 percent of the 16 year old age group.
A single exam for all: the General Certificate of Secondary Education (GCSEs)
The move to a single examination system for all came during a period of economic crisis in the mid-1970s and the 1980s. The slump and rising levels of unemployment produced renewed pressure for education to be as closely geared as possible to the needs of the labour market. The service sector was growing, and manufacturing and manual jobs were in decline; capital required a much higher general level of literacy and numeracy from the workforce to meet the demands of the growing service sector. As a response, the divided examination system of GCSE O levels and CSEs was replaced with a single examination. In 1988, when GCSEs—a single examination for all at 16—were introduced, it was assumed that around 40 percent of pupils would continue to leave school without a certificate. Nevertheless, just 25 years later, hardly any did.
GCSE arrived as part of the Thatcher government’s 1988 Education Reform Act. The new examination failed to meet the aspirations of the comprehensive movement, which had fought for continuous assessment and a single unifying examination for all. GCSEs were a system of subject examinations rather than a single examination with differentiated papers and questions within each subject. There would be seven grades, A to G, and a clear differentiation between those candidates entering for the higher grades (A to C) and those entering for the lower grades (D to G) was built into the system from the start. GCSEs are neither criteria-referenced nor norm-referenced, although exam boards do release grade descriptors. Instead, GCSEs and A levels (taken at 18) use prior attainment referencing, that is, GCSEs are referenced to what the cohort achieved at age 11 and A levels are referenced to what the cohort achieved at GCSE level.41
The ideological driver behind the Education Reform Act of 1988 was the introduction of free market principles into education. Education and exams were to change to reflect the priorities of an increasingly competitive world. The Act centralised control of the curriculum through the introduction of the National Curriculum and introduced assessments for children at 7, 11, 14 and 16 years old.42 Control of the curriculum and the new examinations now rested firmly with central government. Schools were to be ever more closely driven by the concerns of business and the needs of the economy.
The different types of school multiplied from 1988 as the government sought to subject the structure of schooling to market forces and increase the influences of ideas such as competition, privatisation and efficiency. The education system was already a complicated patchwork of private, comprehensive, secondary modern and voluntary-aided schools (as well as a few remaining grammar schools). Soon, foundation, academy and free schools were added, as well as multi-academy trust schools.
Since the school leaving age was raised to 18 in 2015, selection at 11 years old through the eleven-plus has now effectively been effectively replaced by selection at 16 through GCSE grades. Schools are allowed and even encouraged to have highly selective and small academic sixth forms offering mainly A levels while Further Education and Sixth Form Colleges are left to educate the rest.43
GCSEs and prior attainment referencing
Keith Joseph, Margaret Thatcher’s education secretary prior to the introduction of GCSEs, had argued for a “move towards a greater degree of criterion referencing in examinations and away from norm referencing”.44 This was probably a recognition that too many children were being failed by the education system, which was linked in the minds of some commentators to Britain’s poor economic performance. Yet, despite Joseph’s expressed desire, the move ended up being towards the allocation of grades based on neither criterion nor norm reference but rather prior attainment. This involves two processes: ranking and referencing.
Ranking is a process of competitive selection: getting students to compete in the acquisition of skills and knowledge. This requires attaching a number to the skill or an amount of knowledge. This is particularly difficult when it comes to abstract concepts. Take the skill of “problem solving” in maths. First, the skill has to be reified—turned from an abstract concept into something concrete. Once it is concrete, it can be measured and quantified, and this quantification allows students’ “problem solving skills” to be ranked. So, “problem solving” is reified to be considered real, quantified to be understood and ranked to become useful. Reify, quantify, rank: this is the ideology of ranking.45 The rank itself then takes on the feeling of an objective—even “scientific”—measure, which means that ranking seems both inevitable and accurate.
Ranking means that some students will be at the bottom of the rankings by design. Failure is built into the system. Once students are ranked, a decision then needs to be made about the allocation of grades. GCSEs are referenced to a cohort’s prior attainment at Key Stage 2 SATs, that is, what a cohort achieved five years before they take GCSEs.46 Exam boards use this prior attainment data to predict the percentage of students likely to be awarded each grade in a subject, and then this information is used to set grade boundaries in the ranked list of candidates. The result is that grade distributions are roughly similar from year to year and national results remain steady with, for example, an annual failure rate (those achieving grade 3 or below) for GCSE maths of around 30 percent. This group of students are now being referred to as the “forgotten third”.47
There are profound implications of using prior attainment data to set the percentage of students allowed to achieve each grade. Yes, individual schools can improve their results, but only at the expense of other schools’ results going down. Any national attempt to improve, for example, maths or English teaching would not show up because the grade boundaries would simply be moved to ensure that the national results remain in line with prior achievement. The underlying, paradoxical belief thus appears to be that education makes little difference to a student’s educational ability. After five years of secondary education, at the end of year 11, the distribution of students’ abilities remains just as it was at the end of year 6.
This is all very reminiscent of Burt’s ideas of a measurable and unchanging intelligence. His work on intelligence may have been discredited and proved fraudulent in the 1970s, but his more profound mistake of seeing intelligence as measurable and unchangeable often goes unchallenged. Evolutionary biologist and historian of science Steven Jay Gould referred to Burt’s first central error as “reification”: “the notion that such a nebulous, socially defined concept as intelligence might be identified as a ‘thing’ with a locus in the brain…and that it might be measured as a single number, thus permitting a unilinear ranking of people according to the amount of it they possess”.48 Burt’s second big mistake, according to Gould, was that this single number measures an inborn quality of genetic constitution that is highly heritable across generations. His last error was the claim that “a person’s IQ score must be stable and permanent—subject to little change (but only minor and temporary tinkering) by any program of social and educational intervention”.49
Intelligence tests were never intended to be used in the way that Burt deployed them. The tests were invented by French psychologist Alfred Binet during the first decade of the 20th century. Binet was commissioned by France’s ministry of education to identify children who needed some form of special education. Binet explicitly denied that his tests could measure an internal biological property worthy of the name “general intelligence” and rejected any hereditarian readings of his results. In defence of his remedial educational programmes, he insisted that any gains must be read as genuine increases in intelligence:
It is in this practical sense, the only one acceptable to us, that we say that the intelligence of these children has been increased. We have increased what constitutes the intelligence of a pupil: their capacity to learn and their ability to assimilate instruction.50
The use of prior achievement referencing in GCSEs means any increases in ability or intelligence are ruled out in advance. Again, the underlying assumption is that education makes little difference to a student’s educational ability. This should be regarded as a clear sign of failure within the educational system, yet it is built into today’s schooling, and this situation is justified with a battery of spurious arguments: “If the cohort has not changed much, then do not expect the pass rate to change much either”.51
Benjamin Bloom, an American educational psychologist, demonstrated empirically that students actually become better able to learn as a result of the experience of mastery.52 He suggested that a normal distribution of exam results was a sign of educational failure, merely reproducing what nature provided. The job of the teacher was to destroy the bell curve: if some students needed more instructional input to become successful, then it was the teacher’s job to provide it. One of his most important works is a study of stability and change in human characteristics.53 In it he rejects stability as a manifestation of genetically determined factors and concludes that such determinism could be undermined by effective teaching.54
Assigning grades by algorithm
Before the Covid-19 pandemic, ranking and referencing of A level and GCSE grades to prior achievement was carried out behind the scenes and went largely unnoticed. However, with the cancellation of examinations in the summer of 2020, teachers were asked to rank students and assign them grades. Grades were then allocated by an algorithm that essentially ignored the grades awarded by teachers and worked only with the rankings and referencing of prior achievement. It was this that produced such appalling results in 2020.55 Criticism was brushed aside by Boris Johnson, who pointed the finger at a “mutant algorithm”.56 However, the algorithm did exactly what it was meant to do and was—at least initially—the Tories’ preferred approach precisely because it was the closest approximation to business as usual.
This disastrous attitude led to some huge injustices, with marks differing significantly from teacher-assessed grades, and almost 40 percent of A level students received lower than expected grades.57 Widespread protests followed the publication of grades in August 2020.58 The government was forced into a humiliating U-turn, and the algorithm was eventually abandoned and replaced by teacher-assessed grades.59 Had it not been for the protests, many students would have been left with grades that bore no relation to their ability or the work they had done. Having got their fingers burnt in 2020, the government made no attempt to use an algorithm again in 2021, and grades were instead awarded by teachers. The mainstream media complained of grade inflation, and the discussion in government turned to the re-introduction of exams and how they could return to the zero-grade inflation situation of the years prior to Covid-19.
Do we need exams?
Writing in 1948, Jean Piaget, the Swiss psychologist famed for his work on child development, set out his view on examinations:
Everything has been said about the value of scholastic examinations, and yet this veritable plague on education at all levels continues to poison—such terminology is not too strong here—normal relations between the teacher and the student by jeopardising for both parties the joy in work as well as mutual confidence”.60
Since 1948 this “veritable plague” has spread, and now the examination system has an all-pervasive influence on the education system and occupies a position of unrivalled dominance. Not only does it technically fulfil the selective functions required by universities and employers; it also ensures, by means of competition, that people accept the results of selection and has the advantage of being seen as a fair and “scientific” basis on which to select.
Yet, there is a fundamental contradiction encapsulated in the very nature of capitalist education itself. On the one hand, students are being educated to be the next generation of workers and to know their place in society; on the other, education contains within it the power to be a liberating force. It is this capacity for liberation that motivates many teachers. In spite of the exam treadmill, teachers and students manage to carve out a space for education to be as rewarding and fulfilling as possible. Pupils and parents rightly believe that education is a creator of opportunity and a crucial determinant of occupational success. Examinations are seen in their “gatekeeping” role, opening and closing doors to future life chances.
Yet, it is not difficult to imagine education without all this. Instead, we could have a system that prizes the ability to cooperate and values the contribution of even the least able student; a system where the only motivation is found in the subject itself, not passing the exam attached to it. In 1999, after the death of Bloom, academic Elliot Eisner wrote on his contribution to the thinking of educationalists:
His message to the educational world is to focus on attainment and abandon the “horse race” model of schooling that has as its major aim the identification of those who are swiftest. Speed is not the issue, achievement and mastery are. It is that model that should be employed in trying to develop educational programmes for the young.61
Marx argues that the dominant ideas in any society are the ideas of the ruling class.62 Naturally, those who run our society seek to impose their own ideas upon the population through the education of future generations. Rulers naturally tend to ensure that education serves their interests. Examinations—with their emphasis on competition, grading and failure—are an important way of ensuring their interests are met.
Exams and grades may complement the free market but they do not always ensure that workers have the education that employers need. In the wake of the pandemic there are growing calls to scrap GCSEs and move away from exams based on memory and repetition in order to embrace creativity and problem solving.63 Certainly scrapping examinations at 16 and entitling all to a broad and balanced curriculum up to the age of 18 would be a welcome and progressive reform. However, more fundamental reforms are only likely to come alongside a challenge to the values and priorities of capitalist society. If the purpose of education is to liberate human potential, then to fulfil that purpose will require a different economic system—one that is free from competition and exploitation.
Nick Moore is a long-standing member of the Socialist Workers Party in North London. He teaches mathematics at a sixth form college.
1 Thanks to Jane Bassett, Joseph Choonara, Judy Cox, Jai David and Sheila McGregor for comments on an earlier draft.
2 SATs—the “standard attainment tests”—taken in year 6 in primary schools were also cancelled in 2020 and 2021. The General Certificate of Secondary Education and A level are the main school qualifications for 16 and 18 year old students in England, Wales and Northern Ireland.
3 I focus on school exams in England, but many of the same features can be found in other systems. I will not address the history of vocational qualifications, which would need its own analysis.
5 Jinman, 2015. Skills and education are not the same. Education does not necessarily lead to skills and is not necessarily required to develop skills.
6 Noss, 1979, p10.
7 Older systems of schooling were accessible predominantly to the ruling class.
8 For an account on Marx’s theory of exploitation, see Marx, 1976, and Choonara, 2017.
9 Banding is a system in which school pupils are grouped into broad ability bands. Setting is a form of dividing pupils into groups (“sets”) for subjects based on their ability in those subjects.
10 Social Mobility Commission, 2019.
11 Rosenberg, 1991, p14.
12 Bowles and Gintis, 1976.
13 For discussion of problematic determinist readings of this correspondence theory, see Cole, 2008.
14 For Marx, the alienation of labour arises from capitalism. Capitalism transforms labour itself into a commodity; it is taken away from the worker and put under the control of others. Workers are denied expression through their work, and this process has enormous consequences for society. For an extended discussion of alienation, see Swain, 2012.
15 Bourdieu and Passeron, 1977. For a Marxist critique of the concept of “cultural capital”, see Desan, 2013.
16 Broadfoot, 2012, p31.
17 Simon, 1980, p115; Bourne and MacArthur, 1970, p21. The National Union of Elementary Teachers, a forerunner of today’s National Education Union, fought against “payment by results”.
18 Brooks, 2008, p6.
19 Brooks, 2008, p6.
20 Broadfoot, 2012, p37.
21 Morrell, 1989, p79.
22 Brooks, 2008, p1.
23 For a description of the rise of the new middle class, see Callinicos and Harman, 1987, pp21-22.
24 Simon, 1978, pp239-240.
25 Broadfoot, 2012, p43.
26 Rose, Lewontin and Kamin, 1990.
27 Broadfoot, 2012, p44.
28 Simon, 1999, p61.
29 These were the “ordinary” and “advanced” General Certificates of Education.
30 Morrell, 1989, p79.
31 Simon, 1999, p304.
32 Quoted in Noss, 1977.
33 Broadfoot, 2012, p47.
34 Gould, 1981, p264.
35 Morrell, 1989, pp67-68.
37 Coard argues that the cultural biases crystallised in attitudes to the linguistic differences between West Indian English and “standard classroom” English—Coard, 1971.
38 Simon, 1999, p454.
39 Simon, 1999, p483.
40 Morell, 1989, p79.
41 GCSEs were reformed in 2017 with a new grading system. Marks run from 9-1, and grades 8-1 use prior attainment referencing. The top grade 9 is effectively norm-referenced and only available for a small percentage of the cohort—Jadhav, 2017.
42 The National Curriculum only applied to state schools, not to public and private schools.
43 Playfair, 2020, pp45-47. Over 60 percent of all 16-18 year old students in England are studying in colleges.
44 Newton, 2011, p23.
45 Horvath and Bott, 2021, pp22-27.
46 A levels are referenced to prior attainment at GCSE.
47 Association of School and College Leaders, 2019. The effect of failure on a student’s ability, self-confidence and motivation to learn can be profound. One need look no further than the classes of demoralised GCSE students forced to retake maths—students who have been graded as failures after 11 years in the school system—to see the detrimental effect on a student’s motivation and engagement.
48 Gould, 1981, p269.
49 Gould, 1981, p385.
50 Gould, 1981, p388.
51 Newton, 2011, p23.
52 Bloom, 1968.
53 Bloom, 1964.
54 The idea that intelligence is innate and unchangeable resurfaced with the publication of The Bell Curve: Intelligence and Class Structure in American Life by Richard Herrnstein and Charles Murray in the 1990s. It still has currency today within ruling-class circles, as evidenced by Dominic Cummings’s claim that up to 70 percent of a child’s performance is related to their genes—Helm, 2013.
55 For an explanation of the technicalities of the A level algorithm, see Hern, 2020.
56 Stewart, 2020.
57 Duncan, McIntyre, Storer and Levett, 2020. Individual examples from my own experience include an AS level student who was awarded a grade U by the algorithm despite receiving a grade C from their teacher.
58 Braddick, 2020.
59 The government settled on awarding students the better of the teacher-assessed or algorithm grades.
60 Piaget, 1973, p73.
61 Eisner, 2000. Bloom’s “Learning for Mastery” contains many ideas that could be used in the construction of an alternative education system—Bloom, 1968.
62 Marx, 1974.
63 Carr, 2021.