Want to judge people on merit? Then ignore their exam results

Ed Smith's "Left Field" column.

Students sitting exams.
Students sitting exams. Photograph: Getty Images

What do we mean by the phrases “good education” and “an educated person”? It is a surprisingly difficult and revealing question. Is it a set of academic credentials? Or does it mean preparedness for the workplace, someone who is likely to get a good job? Or does is it imply something much broader and harder to define – intellectual curiosity and an attitude of mind?

Having visited several secondary schools this month in very different parts of the country, I repeatedly heard teachers complain about the British exam system. Some of their criticisms were familiar, others quite new.

It now widely accepted that British pupils are excessively over-examined. Teachers are so busy focusing on examinations that there is little time left for education. Exam-led cramming has become the year-round norm – like an election campaign that consumes the whole political cycle. Exams are obviously necessary. But there is an optimal amount of assessment and it has been far exceeded. Grade inflation – notwithstanding this year’s controversial “crackdown” – is simply accepted as a fact.

Ticked off

Other side effects of over-examination are less well-known. There are now so many exam scripts to process that exam boards can’t afford to pay people properly to mark the papers. So the number of experienced, reliable examiners has been reduced. Furthermore, given the prescriptive rules of “grid marking”, it is a risk for pupils to engage too intellectually. One teacher told me with regret that she had to advise her most academic pupil not to display her full intellectual range: in order to secure all the ticks, first you have to stop thinking freely.

As the standard of marking has declined, grades are awarded with increasing randomness. It is an unfortunate irony: in our age of credentialism, exams have never mattered more. And yet they have never been more unreliable as gauges of academic quality.

There is a much deeper problem with exam-dominated education. Credentialism has not continued to expand opportunity. The more closely universities select school leavers according to their exam grades – and the more strictly employers award jobs to people with the best degrees – the more likely it is that savvy middle-class parents will find a way to play the system according to their children’s interests.

Given the lengths these parents will go to in order to engineer superb grades for their children – private tutors, cramming, re-marks, re-sits – a bald set of results on a piece of paper now means remarkably little. The correlation between exam results and intelligence has been steadily weakening. Coursework, for example, was supposed to reduce the stress of a daunting exam season. In reality, it has worked in favour of pupils with pushy parents and experienced teachers who know how to push the boundaries of “internal assessment”.

Credentialism was supposed to be a fairer alternative to the winks and nudges of the old-boy network. But the idea has turned full circle. It now plays into the hands of people who have the energy and inside knowledge to navigate the system. Entrenched interests are always the fastest to adapt to new rules of social and economic advancement.

David Willetts made a similar point in his essay The Future of Meritocracy in 2006: “Over the past 20 years, Britain has seen a big increase in the earnings of graduates relative to non-graduates. It does indeed look as though the expansion in higher education has meant, above all, more places for students from more affluent backgrounds rather than students from poorer backgrounds . . . This means that the expansion of higher education has not increased social mobility but, if anything, has contributed to its decline.”

This uncomfortable argument leads to a counterintuitive solution: potential, not just performance, should play a greater role in the distribution of university places (and jobs). Sadly, only a few educational institutions, such as Harvard, have the conviction and financial resources to follow this logic to its conclusion. The Harvard admissions process is famously dismissive of high-school grades, placing more emphasis on its own interview system. When he was invited to observe the Harvard admissions process, Peter Lampl (who founded the educational charity the Sutton Trust) was struck by how Harvard awarded places according to character and potential rather than pure exam results. When he asked why one girl was admitted despite her relatively poor grades, the tutor for admissions replied: “We think she will do great things, perhaps become the mayor of a major city.” Harvard focused on the contribution the undergraduate would make – not just at university but in civic society.

Brown nosy

Ironically, Gordon Brown’s infamous intervention in the Laura Spence affair in 1999 grasped exactly the wrong end of the stick. Brown objected to the idea that a state-educated candidate could have flawless academic grades yet still be rejected by Magdalen College, Oxford. Whatever the rights and wrongs of that particular example, Brown would have better advanced the meritocratic agenda by advocating a general loosening of the link between exam grades and university admissions. On average – given how effectively private schools deliver impeccable exam results – pupils such as Spence would have more chance of getting in to Oxbridge colleges if academic grades carried less weight, not more.

Trying vainly to turn education into an exact science that can be measured precisely does not make it fairer. Instead, it reduces the scope for far-sighted universities and employers to back their judgement.

There is a further dimension to the problem of credentialism. It encourages life’s winners to underestimate their good fortune and to overrate the extent to which they deserve their success. Far from advancing talent over privilege, credentialism has strengthened the grip of people already at the top.

Ed Smith’s “Luck: What It Means and Why It Matters” is published by Bloomsbury (£16.99)