Discuss this article at First Drafts,Prospect'sblog
The familiar sounds of an early English summer are with us once again. Millions of children sit down to Sats, GCSEs, AS-levels, A-levels and a host of lesser exams, and the argument over educational standards starts. Depending on whom you listen to, we should either be letting up on over-examined pupils by abolishing Sats, and even GCSEs, or else making exams far more rigorous.
The chorus will reach a crescendo when GCSE and A-level results are published in August. If pass rates rise again, commentators will say that standards are falling because exams are getting easier. If pass rates drop, they will say that standards are falling because children are getting lower marks. Parents like myself try to ignore this and base our judgements on what our children are learning. But it's not easy given how much education has changed since we were at school.
Some trends are encouraging—education has been made more relevant and enthuses many children that it would have previously bored. My sons' A-level French revision involved listening to radio debates on current affairs, whereas mine involved rereading Molière. And among their peers, a far greater proportion stayed in education for longer. In the good comprehensive my children attended, I would guess that two thirds of the year group were still learning at the age of 17. In the good comprehensive I attended in the 1970s, I reckon that two thirds had either left school by the same age, or were enrolled but did virtually no work.
On the other hand, some aspects of schooling today are incomprehensible to my generation, such as gaps in general knowledge (it's a while since I met a 17 year old with a clear picture of how events unfolded in the 20th century) and the hand-holding that goes with ensuring that students leave with good grades. My experience has been that even when we parents resist the temptation to help with GCSE or A-level coursework, a teacher with the child's interests at heart may send a draft piece of work back several times with pointers to how it can be improved before the examiners see it.
The debate about standards persists because there is no single objective answer to the question "are standards better or worse than they were a generation ago"? Each side points to indicators that favour them, in the knowledge that there is no authoritative definition, let alone a measure that has been consistently applied over the decades. But the annual soul-searching over exams is about more than student assessment. It reveals a national insecurity about whether our education system is teaching the right things. It is also fed by an anxiety about whether, in a country with a history of upholding standards by ensuring that plenty of students fail, we can attain the more modern objective of ensuring that every child leaves school with something to show for it. The longer I spent looking at what the tests and exam results tell us, the more I was drawn to a fundamental question: what kind of education do we want for our children?
Measuring with an elastic ruler If exam results and the completion of stages of education are taken as a consistent unit of measurement, standards have soared in the past generation. In 1980, the year I graduated from university, about one in ten people of my age got a degree. For the generation of my son, who is now an undergraduate, it will be close to half. When I sat my O-levels, about one in three of my contemporaries got qualifications enabling them to attend sixth form; today nearly two in three do so.
In 2004, an independent committee headed by Barry McGaw, an Australian who was head of education at the OECD, looked at whether A-level standards were being maintained. His conclusion was that this is the wrong question. Exams, he said, are best designed for purposes other than monitoring standards over time. The main function of A-levels is to allow universities to distinguish between students in making admissions, and to ensure that the right things are being taught in schools. British exam boards are doing as well as anywhere in ensuring that standards are consistent, but it is futile to try to keep the meaning of a grade the same over long periods of time. In another review in 2008, commissioned by the Office of the Examinations Regulator, none of the experts consulted could suggest a way of keeping standards consistent over a long period, and chose instead to focus on consistency across subjects and between different boards.
McGaw's most significant recommendation was not addressed to the exam boards or government but to the media and public. Change what you ask of the exam system, he said. "It would help if different expectations were set, not asking if performance standards are rising or falling over the long term but asking only if the examinations are making reasonable and appropriate demands of students." Fat chance.
Yet, for qualifications to be useful, society must understand what they represent. This is not much of a problem for universities, who use A-levels as a form of currency to select the best applicants. If grade inflation means that a B today is equivalent to a C yesterday, that does not matter any more than adding a zero onto a currency: the price of any given place simply goes up. The real difficulty arises when—as has occurred—one in four candidates have As, and universities cannot find the brightest students. This is then solved by subdividing the highest grade, one of the reasons why an A* grade will be awarded at A-level from next year.
So it does not matter much to universities if grades do not hold their perceived value over time. But it matters more to employers, who want assurance that a qualification signifies certain levels of competence, and who recruit from a pool of people who have obtained their qualifications at different times. They need broadly recognisable benchmarks. I put this point to Harvey Goldstein, an expert in educational measurement, and one of the many purists who think you can't find a meaningful measure of standards over time. "Employers only ever see a partial picture," he said "depending on the individuals they encounter." He accepts that too much variation is unhelpful, but reckons that employers have never accurately understood what an educational qualification really certifies.
More importantly, though, Goldstein and others I talked to agreed that exam standards are at least partially influenced by assumptions about what kind of results they should produce. Gone are the days when A-level grades were calibrated to ensure that roughly the same number pass each year at each grade. However, it is far from clear that this has been replaced by objective judgements about what a given grade represents. Examiners are under pressure to substantiate a belief—held not just by the government but by much of the educational establishment—that teaching and learning are improving over time. There is therefore a tendency to set levels that allow more children to pass. Each year, examiners have to make decisions about where to put "cut points"—the scores that distinguish one exam grade from another. And they are just as much at risk of being criticised for being too severe as for being too lenient. Remember the s candal in 2002 when A-level examiners arbitrarily marked students down because they feared criticism over grade inflation? It cost the head of the exams watchdog his job and contributed to the resignation of education secretary Estelle Morris.
Moreover, anyone who sits GCSEs or A-levels is urged to look closely at marking schemes to see how best to obtain points on particular questions. This is more feasible than in the past not just because of the publication of marking schemes but also because of the breaking down of tasks into short questions—in many cases, replacing essay questions. Defenders of the system say that it is good that students are given more chances to show what they know, and that the criteria for doing well are no longer secret. Even if you swallow this argument—and there must be a point where hand-holding and second chances go too far—it's clear that the route to an A grade is easier than it once was.
A bigger, better middle? Yet none of this shows that the overall standard being achieved by students has slipped. Even if we agreed, say, that an A at A-level today is equivalent to a B a generation ago, or that degrees at some universities are worth little more than a good school-leaving qualification in the past, the overall educational profile of the population would have improved. Slightly more young people are getting degrees today than even managed to get decent O-levels when I was at school.
In particular, the educational experience of young people in the middle of the ability range has been transformed. Large numbers are being educated to age 18 or 21 who in the past would have left with few or no qualifications at 15 or 16. This must in part be positive news. For example, six in ten 16 year olds now get a GCSE at grade C or above in maths. Thirty years ago, most young people were turned off maths long before that age. Even if a grade C in maths GCSE is not that demanding, most 16 year olds are at least getting a qualification—helping to combat the "I can't do maths" syndrome that hampers so many British adults.
Potentially, some of the limitations with measuring standards through exam results can be overcome by using surveys that test the knowledge and skills of a sample of the population. Even here, it is not easy to make comparisons over time (purists would say it is impossible), but some of the evidence is useful. The idea that the population as a whole is becoming more intelligent is supported by the "Flynn effect," named after James Flynn, a New Zealand academic specialising in intelligence measurement. Flynn showed a consistent rise in global IQ performance of roughly three percentage points a decade, in some cases going back to the early 20th century (although possibly levelling off in the 1990s in some rich countries). While the meaning of the Flynn effect can be debated, an increase in the quantity of education has therefore been accompanied by a steady rise in performance on the most consistent measure of thinking ability.
A 2008 report by Michael Shayer of Kings College, London appeared to contradict this evidence by showing a sharp deterioration over 30 years in the ability of the most able students to think logically. In fact the study's less prominently reported finding was that the average score in this test had improved. It was the numbers getting top marks that had fallen. This is consistent with the Flynn effect to the extent that overall improvement has been driven to a large degree by rises in the IQ scores of the less able. It makes sense given that it is young people of less advantaged backgrounds, in the past largely written off after primary education, who have benefited most from the expansion of educational participation. But it does raise the issue of whether this has been at the expense of the most able.
Another set of evidence comes from international surveys in which Britain's performance has been middling to good. These surveys do not yet provide usable comparisons over a very long period, but the latest results from one, the Trends in International Mathematics and Science Study (TIMSS), appear to show substantial improvements since 1995 among English students aged ten and 14 in maths. I have learned to be cautious about interpreting such results through my involvement in the Programme for International Student Assessment (Pisa), a study of the knowledge and skills of 15 year olds. In the first round of that study, in 2000, Britain appeared to do very well, but it was later found that this was probably an inaccurate sample. The main rule in interpreting these surveys is patience: the big ones that began to measure student performance in the mid-1990s are only just starting to record significant change in any country's performance.
However, such studies can already be used to help to corroborate or contradict the evidence of our national tests such as Sats. A prominent former government adviser told me that the progress of some of these test results often closely tracked the effects of government initiatives. After an initial push on literacy standards in the late 1990s, not only did the Sats results in primary school English soar, but by 2001 England's ten year olds were ranked third on the PIRLS international reading test. Thereafter he said there was a "loss of focus" in government on primary school literacy, the Sats results levelled off, and by 2006 England had slipped to 15th place in the PIRLS test. By then, however, a second push to raise standards on numeracy was soon followed by excellent maths results for ten year olds in the 2007 TIMSS test. This at least suggests that hyperactivity in Whitehall, while frequently criticised, can have an impact measurable in independent tests.
How to challenge young minds It seems, then, that the standard achieved at least by the bulk of young people in the middle of the ability range has probably risen. But are the knowledge and skills they are acquiring appropriate for their future? Put simply: have we been right to dump didactic teaching methods? And will middle achievers benefit from extra years of study?
There is no shortage of critics of today's teaching methods. In their vanguard is Chris Woodhead, former chief inspector of schools and scourge of the progressive establishment. For him, the idea that what matters in education is not knowledge, but skills has "emptied the national curriculum of any worthwhile significance." By this he means that we have acquired a disdain for facts and the direct transmission of knowledge from teacher to student to the point that students lack any meaningful understanding of subjects. Traditionalists are highly sceptical of modern teaching methods that emphasise the role of the teacher as a "facilitator" of student inquiry rather than the passing on of a body of knowledge.
There is in fact less that separates the approaches of Woodhead and the progressive teaching establishment than either would admit. Woodhead is careful to distinguish himself from the stereotype of Gradgrind—who saw children as like pitchers to be filled to the brim with facts. Woodhead says children should be taught how to think, but that knowledge is a prerequisite to thought. The teachers I have talked to (see "What do teachers think about standards") agree that children need knowledge, but believe that thought and reflection is a prerequisite to acquiring it in a way that will make sense to them.
Another crucial question is whether a reluctance to set students tasks at which they may fail leads to a reduction in ambition, and a loss of rigour. Of course, students themselves can adopt risk-averse strategies. A classic case is A-level languages, which now attract fewer takers because they are more intellectually demanding than some other subjects. My son was in classes of only two for both French and German Nationally, A-level entries in these subjects have halved in a decade. Learning German grammar undoubtedly helps train the mind in a way that a university would welcome, but universities do little to distinguish between harder and easier A-levels in their conditional offers. Until they do, students will realise that opting for easier subjects is in their best interest.
Nor is it always clear that students are being well served by being encouraged to study for longer. In Britain, where until relatively recently the majority of children never reached a level of education that gave them qualifications of value, it is tempting to be caught up in the thinking that more is better. In the mid 1980s, only half of 17 year olds in Britain were enrolled in education, compared to 81 per cent of French and 97 per cent of German youngsters. It is hardly surprising that raising staying-on rates has been seen as an end in itself. Today over three quarters stay on beyond 16, but the legacy of a post-16 system designed for an academic elite remains.
Alan Smithers, professor of education at the University of Buckingham, mounts a powerful challenge to the idea that students in the middle are doing better now by getting more education. "Those in the middle," he says, "may actually be getting a worse deal now, because quite a few are being drawn into university to do things that are not going to enhance their lives very much. In previous times they might have gone into jobs and got useful qualifications."
Smithers's thesis, echoed by other experts and teachers I talked to, notes our tendency to seek respectability by making everything theoretical—whether it is trying to squeeze theory into a photography A-level, putting high-level maths into a technical engineering diploma to get it recognised as a university entrance qualification, or encouraging someone with Cs at GCSE to struggle through A-levels and a general course at a university. When half of the workforce are graduates, a third-rate degree will make little difference in the labour market. Nor does it help to look backwards and restore a sheep-and-goats system, segregating those who work with their brains from those who work with their hands. Many jobs today require a combination of technical and general skills that were not served by giving some pupils a poor education at secondary school and then rescuing them with an apprenticeship or City and Guilds course. Smithers believes that we can solve the problem only if we produce a new set of intermediate qualifications which accepts that technicians do not need university-style theoretical knowledge.
The skills we need As we have seen, the standards debate has too often been a pointless fight between those who think that any improvement is a fix, and those who think that getting more children through tests is a sign that our system is working. A more worthwhile debate would focus on what we want our children to know and do when they emerge from education.
First, we need to resolve whether we want a system in which failure is a central feature and, if not, how we can still give it rigour. An important function of the education system through much of the 20th century was to sort young people; as in The Weakest Link, the losers left with nothing. With only one in seven workers now employed in unskilled occupations, it is reasonable to create a 21st-century education system that in principle allows every child to achieve something.
Much of our difficulty has been in managing this transition. We have been obsessed with academic excellence as the paradigm of success for so long that we are poor at educating for a range of destinations. A first step away from this is to recognise that a simple vocational-academic distinction in education is false: we do not sniff at law schools. In Singapore, the overlap between the theoretical and the applied is recognised in the slogan "we think with our hands."
In this sense, perhaps our biggest mistake was to abolish the distinction between polytechnics and universities. It is clear that we need a variety of institutions and qualifications, and it doesn't help for every place of post-18 learning to call itself a university and try to prove its academic credentials. Academic snobbery will always exist, but France has shown that two-year technical schools and grandes écoles—both highly reputed institutions—can coexist with universities. Sadly, one factor that has helped to reduce demand for a liberal education at a general university is the high rate of failure of weaker students on these courses—both to complete them and to get jobs after graduation. In Britain, stronger former polytechnics such as Nottingham Trent and Wolverhampton are well placed to be hubs in their regions, and do not need to compete on equal terms with Oxbridge.
Creating a something-for-everyone system also relies on finding an effective school curriculum. Even if by the age of 14, 16 or 18 different students are heading in varied directions, everyone still needs to reach a common level of numeracy, literacy and general thinking skills. Succeeding without abandoning rigour may mean maintaining exam systems that are capable of failing people, but also having the confidence to celebrate a low failure rate as a sign of success, rather than of a flawed system.
Finally, we must agree that young people require both knowledge and skills, but most of all they need to learn how to think. Theories about what should be included in the curriculum and the best way to teach will wax and wane, but however we monitor standards in the future we must assess students' ability to use their critical faculties. In the age of the internet, retrieving information is easy; processing and interpreting it is the challenge. These are age-old skills, but ones that people will require more than ever before. The need for high educational standards has never been greater.
Standards: going up or going down?
Going up
Going down
The familiar sounds of an early English summer are with us once again. Millions of children sit down to Sats, GCSEs, AS-levels, A-levels and a host of lesser exams, and the argument over educational standards starts. Depending on whom you listen to, we should either be letting up on over-examined pupils by abolishing Sats, and even GCSEs, or else making exams far more rigorous.
The chorus will reach a crescendo when GCSE and A-level results are published in August. If pass rates rise again, commentators will say that standards are falling because exams are getting easier. If pass rates drop, they will say that standards are falling because children are getting lower marks. Parents like myself try to ignore this and base our judgements on what our children are learning. But it's not easy given how much education has changed since we were at school.
Some trends are encouraging—education has been made more relevant and enthuses many children that it would have previously bored. My sons' A-level French revision involved listening to radio debates on current affairs, whereas mine involved rereading Molière. And among their peers, a far greater proportion stayed in education for longer. In the good comprehensive my children attended, I would guess that two thirds of the year group were still learning at the age of 17. In the good comprehensive I attended in the 1970s, I reckon that two thirds had either left school by the same age, or were enrolled but did virtually no work.
On the other hand, some aspects of schooling today are incomprehensible to my generation, such as gaps in general knowledge (it's a while since I met a 17 year old with a clear picture of how events unfolded in the 20th century) and the hand-holding that goes with ensuring that students leave with good grades. My experience has been that even when we parents resist the temptation to help with GCSE or A-level coursework, a teacher with the child's interests at heart may send a draft piece of work back several times with pointers to how it can be improved before the examiners see it.
The debate about standards persists because there is no single objective answer to the question "are standards better or worse than they were a generation ago"? Each side points to indicators that favour them, in the knowledge that there is no authoritative definition, let alone a measure that has been consistently applied over the decades. But the annual soul-searching over exams is about more than student assessment. It reveals a national insecurity about whether our education system is teaching the right things. It is also fed by an anxiety about whether, in a country with a history of upholding standards by ensuring that plenty of students fail, we can attain the more modern objective of ensuring that every child leaves school with something to show for it. The longer I spent looking at what the tests and exam results tell us, the more I was drawn to a fundamental question: what kind of education do we want for our children?
Measuring with an elastic ruler If exam results and the completion of stages of education are taken as a consistent unit of measurement, standards have soared in the past generation. In 1980, the year I graduated from university, about one in ten people of my age got a degree. For the generation of my son, who is now an undergraduate, it will be close to half. When I sat my O-levels, about one in three of my contemporaries got qualifications enabling them to attend sixth form; today nearly two in three do so.
In 2004, an independent committee headed by Barry McGaw, an Australian who was head of education at the OECD, looked at whether A-level standards were being maintained. His conclusion was that this is the wrong question. Exams, he said, are best designed for purposes other than monitoring standards over time. The main function of A-levels is to allow universities to distinguish between students in making admissions, and to ensure that the right things are being taught in schools. British exam boards are doing as well as anywhere in ensuring that standards are consistent, but it is futile to try to keep the meaning of a grade the same over long periods of time. In another review in 2008, commissioned by the Office of the Examinations Regulator, none of the experts consulted could suggest a way of keeping standards consistent over a long period, and chose instead to focus on consistency across subjects and between different boards.
McGaw's most significant recommendation was not addressed to the exam boards or government but to the media and public. Change what you ask of the exam system, he said. "It would help if different expectations were set, not asking if performance standards are rising or falling over the long term but asking only if the examinations are making reasonable and appropriate demands of students." Fat chance.
Yet, for qualifications to be useful, society must understand what they represent. This is not much of a problem for universities, who use A-levels as a form of currency to select the best applicants. If grade inflation means that a B today is equivalent to a C yesterday, that does not matter any more than adding a zero onto a currency: the price of any given place simply goes up. The real difficulty arises when—as has occurred—one in four candidates have As, and universities cannot find the brightest students. This is then solved by subdividing the highest grade, one of the reasons why an A* grade will be awarded at A-level from next year.
So it does not matter much to universities if grades do not hold their perceived value over time. But it matters more to employers, who want assurance that a qualification signifies certain levels of competence, and who recruit from a pool of people who have obtained their qualifications at different times. They need broadly recognisable benchmarks. I put this point to Harvey Goldstein, an expert in educational measurement, and one of the many purists who think you can't find a meaningful measure of standards over time. "Employers only ever see a partial picture," he said "depending on the individuals they encounter." He accepts that too much variation is unhelpful, but reckons that employers have never accurately understood what an educational qualification really certifies.
More importantly, though, Goldstein and others I talked to agreed that exam standards are at least partially influenced by assumptions about what kind of results they should produce. Gone are the days when A-level grades were calibrated to ensure that roughly the same number pass each year at each grade. However, it is far from clear that this has been replaced by objective judgements about what a given grade represents. Examiners are under pressure to substantiate a belief—held not just by the government but by much of the educational establishment—that teaching and learning are improving over time. There is therefore a tendency to set levels that allow more children to pass. Each year, examiners have to make decisions about where to put "cut points"—the scores that distinguish one exam grade from another. And they are just as much at risk of being criticised for being too severe as for being too lenient. Remember the s candal in 2002 when A-level examiners arbitrarily marked students down because they feared criticism over grade inflation? It cost the head of the exams watchdog his job and contributed to the resignation of education secretary Estelle Morris.
Moreover, anyone who sits GCSEs or A-levels is urged to look closely at marking schemes to see how best to obtain points on particular questions. This is more feasible than in the past not just because of the publication of marking schemes but also because of the breaking down of tasks into short questions—in many cases, replacing essay questions. Defenders of the system say that it is good that students are given more chances to show what they know, and that the criteria for doing well are no longer secret. Even if you swallow this argument—and there must be a point where hand-holding and second chances go too far—it's clear that the route to an A grade is easier than it once was.
A bigger, better middle? Yet none of this shows that the overall standard being achieved by students has slipped. Even if we agreed, say, that an A at A-level today is equivalent to a B a generation ago, or that degrees at some universities are worth little more than a good school-leaving qualification in the past, the overall educational profile of the population would have improved. Slightly more young people are getting degrees today than even managed to get decent O-levels when I was at school.
In particular, the educational experience of young people in the middle of the ability range has been transformed. Large numbers are being educated to age 18 or 21 who in the past would have left with few or no qualifications at 15 or 16. This must in part be positive news. For example, six in ten 16 year olds now get a GCSE at grade C or above in maths. Thirty years ago, most young people were turned off maths long before that age. Even if a grade C in maths GCSE is not that demanding, most 16 year olds are at least getting a qualification—helping to combat the "I can't do maths" syndrome that hampers so many British adults.
Potentially, some of the limitations with measuring standards through exam results can be overcome by using surveys that test the knowledge and skills of a sample of the population. Even here, it is not easy to make comparisons over time (purists would say it is impossible), but some of the evidence is useful. The idea that the population as a whole is becoming more intelligent is supported by the "Flynn effect," named after James Flynn, a New Zealand academic specialising in intelligence measurement. Flynn showed a consistent rise in global IQ performance of roughly three percentage points a decade, in some cases going back to the early 20th century (although possibly levelling off in the 1990s in some rich countries). While the meaning of the Flynn effect can be debated, an increase in the quantity of education has therefore been accompanied by a steady rise in performance on the most consistent measure of thinking ability.
A 2008 report by Michael Shayer of Kings College, London appeared to contradict this evidence by showing a sharp deterioration over 30 years in the ability of the most able students to think logically. In fact the study's less prominently reported finding was that the average score in this test had improved. It was the numbers getting top marks that had fallen. This is consistent with the Flynn effect to the extent that overall improvement has been driven to a large degree by rises in the IQ scores of the less able. It makes sense given that it is young people of less advantaged backgrounds, in the past largely written off after primary education, who have benefited most from the expansion of educational participation. But it does raise the issue of whether this has been at the expense of the most able.
Another set of evidence comes from international surveys in which Britain's performance has been middling to good. These surveys do not yet provide usable comparisons over a very long period, but the latest results from one, the Trends in International Mathematics and Science Study (TIMSS), appear to show substantial improvements since 1995 among English students aged ten and 14 in maths. I have learned to be cautious about interpreting such results through my involvement in the Programme for International Student Assessment (Pisa), a study of the knowledge and skills of 15 year olds. In the first round of that study, in 2000, Britain appeared to do very well, but it was later found that this was probably an inaccurate sample. The main rule in interpreting these surveys is patience: the big ones that began to measure student performance in the mid-1990s are only just starting to record significant change in any country's performance.
However, such studies can already be used to help to corroborate or contradict the evidence of our national tests such as Sats. A prominent former government adviser told me that the progress of some of these test results often closely tracked the effects of government initiatives. After an initial push on literacy standards in the late 1990s, not only did the Sats results in primary school English soar, but by 2001 England's ten year olds were ranked third on the PIRLS international reading test. Thereafter he said there was a "loss of focus" in government on primary school literacy, the Sats results levelled off, and by 2006 England had slipped to 15th place in the PIRLS test. By then, however, a second push to raise standards on numeracy was soon followed by excellent maths results for ten year olds in the 2007 TIMSS test. This at least suggests that hyperactivity in Whitehall, while frequently criticised, can have an impact measurable in independent tests.
How to challenge young minds It seems, then, that the standard achieved at least by the bulk of young people in the middle of the ability range has probably risen. But are the knowledge and skills they are acquiring appropriate for their future? Put simply: have we been right to dump didactic teaching methods? And will middle achievers benefit from extra years of study?
There is no shortage of critics of today's teaching methods. In their vanguard is Chris Woodhead, former chief inspector of schools and scourge of the progressive establishment. For him, the idea that what matters in education is not knowledge, but skills has "emptied the national curriculum of any worthwhile significance." By this he means that we have acquired a disdain for facts and the direct transmission of knowledge from teacher to student to the point that students lack any meaningful understanding of subjects. Traditionalists are highly sceptical of modern teaching methods that emphasise the role of the teacher as a "facilitator" of student inquiry rather than the passing on of a body of knowledge.
There is in fact less that separates the approaches of Woodhead and the progressive teaching establishment than either would admit. Woodhead is careful to distinguish himself from the stereotype of Gradgrind—who saw children as like pitchers to be filled to the brim with facts. Woodhead says children should be taught how to think, but that knowledge is a prerequisite to thought. The teachers I have talked to (see "What do teachers think about standards") agree that children need knowledge, but believe that thought and reflection is a prerequisite to acquiring it in a way that will make sense to them.
Another crucial question is whether a reluctance to set students tasks at which they may fail leads to a reduction in ambition, and a loss of rigour. Of course, students themselves can adopt risk-averse strategies. A classic case is A-level languages, which now attract fewer takers because they are more intellectually demanding than some other subjects. My son was in classes of only two for both French and German Nationally, A-level entries in these subjects have halved in a decade. Learning German grammar undoubtedly helps train the mind in a way that a university would welcome, but universities do little to distinguish between harder and easier A-levels in their conditional offers. Until they do, students will realise that opting for easier subjects is in their best interest.
Nor is it always clear that students are being well served by being encouraged to study for longer. In Britain, where until relatively recently the majority of children never reached a level of education that gave them qualifications of value, it is tempting to be caught up in the thinking that more is better. In the mid 1980s, only half of 17 year olds in Britain were enrolled in education, compared to 81 per cent of French and 97 per cent of German youngsters. It is hardly surprising that raising staying-on rates has been seen as an end in itself. Today over three quarters stay on beyond 16, but the legacy of a post-16 system designed for an academic elite remains.
Alan Smithers, professor of education at the University of Buckingham, mounts a powerful challenge to the idea that students in the middle are doing better now by getting more education. "Those in the middle," he says, "may actually be getting a worse deal now, because quite a few are being drawn into university to do things that are not going to enhance their lives very much. In previous times they might have gone into jobs and got useful qualifications."
Smithers's thesis, echoed by other experts and teachers I talked to, notes our tendency to seek respectability by making everything theoretical—whether it is trying to squeeze theory into a photography A-level, putting high-level maths into a technical engineering diploma to get it recognised as a university entrance qualification, or encouraging someone with Cs at GCSE to struggle through A-levels and a general course at a university. When half of the workforce are graduates, a third-rate degree will make little difference in the labour market. Nor does it help to look backwards and restore a sheep-and-goats system, segregating those who work with their brains from those who work with their hands. Many jobs today require a combination of technical and general skills that were not served by giving some pupils a poor education at secondary school and then rescuing them with an apprenticeship or City and Guilds course. Smithers believes that we can solve the problem only if we produce a new set of intermediate qualifications which accepts that technicians do not need university-style theoretical knowledge.
The skills we need As we have seen, the standards debate has too often been a pointless fight between those who think that any improvement is a fix, and those who think that getting more children through tests is a sign that our system is working. A more worthwhile debate would focus on what we want our children to know and do when they emerge from education.
First, we need to resolve whether we want a system in which failure is a central feature and, if not, how we can still give it rigour. An important function of the education system through much of the 20th century was to sort young people; as in The Weakest Link, the losers left with nothing. With only one in seven workers now employed in unskilled occupations, it is reasonable to create a 21st-century education system that in principle allows every child to achieve something.
Much of our difficulty has been in managing this transition. We have been obsessed with academic excellence as the paradigm of success for so long that we are poor at educating for a range of destinations. A first step away from this is to recognise that a simple vocational-academic distinction in education is false: we do not sniff at law schools. In Singapore, the overlap between the theoretical and the applied is recognised in the slogan "we think with our hands."
In this sense, perhaps our biggest mistake was to abolish the distinction between polytechnics and universities. It is clear that we need a variety of institutions and qualifications, and it doesn't help for every place of post-18 learning to call itself a university and try to prove its academic credentials. Academic snobbery will always exist, but France has shown that two-year technical schools and grandes écoles—both highly reputed institutions—can coexist with universities. Sadly, one factor that has helped to reduce demand for a liberal education at a general university is the high rate of failure of weaker students on these courses—both to complete them and to get jobs after graduation. In Britain, stronger former polytechnics such as Nottingham Trent and Wolverhampton are well placed to be hubs in their regions, and do not need to compete on equal terms with Oxbridge.
Creating a something-for-everyone system also relies on finding an effective school curriculum. Even if by the age of 14, 16 or 18 different students are heading in varied directions, everyone still needs to reach a common level of numeracy, literacy and general thinking skills. Succeeding without abandoning rigour may mean maintaining exam systems that are capable of failing people, but also having the confidence to celebrate a low failure rate as a sign of success, rather than of a flawed system.
Finally, we must agree that young people require both knowledge and skills, but most of all they need to learn how to think. Theories about what should be included in the curriculum and the best way to teach will wax and wane, but however we monitor standards in the future we must assess students' ability to use their critical faculties. In the age of the internet, retrieving information is easy; processing and interpreting it is the challenge. These are age-old skills, but ones that people will require more than ever before. The need for high educational standards has never been greater.
Standards: going up or going down?
Going up
- 1.5m: the number of women studying at British universities today, up from 300,000 in 1980. The number of men has risen from 500,000 to 1.1m.
- 64.8 per cent of 15 year olds gain five GCSEs at grade A* to C, an increase from 43.5 per cent since 1994-95.
- 26 per cent of A-level entries received a grade A in 2008, and the pass rate improved for the 28th consecutive year.
Going down
- 1 in 40: the number of English students studying French at AS level. The number of students doing A-level French or German has halved in the past decade. The great majority of French 17 year olds are studying English.
- 88 per cent of 16-19 year olds are in education, employment or training, compared to 90 per cent a decade ago.
- 1 in 9 13 and 14 year olds got high marks on logical questions requiring deductions about a pendulum on a string, compared to 1 in 4 on the same questions ?in 1976.
