In fact, a considerable body of evidence suggests that exams are not even the most useful way to assess pupils - that is if we are talking about education and not accountability. And here’s the rub: in the current system it is not clear why we assess pupils at all.

If we are simply talking about accountability, then exams apparently perform many useful functions. They allow a government to point to rising standards and the opposition to bleat that standards are slipping. On the surface, they also enable politicians to identify successful and failing schools through league tables. These tables, in theory at least, also help parents make an informed choice of school. Most importantly, exams allow the pupils to have a certificate which provides a passport to the next stage in their career, be it school, college or workplace.

If, however, we want to discuss education, the purpose of this summer madness is less obvious.

Exams now so dominate the school system that they are in danger of obscuring all other reasons for being educated. Yet, away from the pressure of high stakes testing, assessment is possibly the single most important teachers’ tool for improving pupils’ performance.

Research by Paul Black and Dylan Wiliam of King’s College London makes this point unequivocally. They reviewed all the work done on “formative” assessment - where, rather than simply marking and grading, pupils’ work is analysed and they are given feedback on how they can improve. Their results were startling - showing an improved performance equivalent to, for example, improving our world ranking in maths from somewhere in the middle to near the top.

You would think a government keen to raise standards would leap at such research and implement it with all the enthusiasm that attended the introduction of literacy and numeracy hours. Not so. For among their findings was the nugget that constant grading and testing actually prevented pupils from improving. The bright became complacent and the struggling got discouraged.

Even comments with grades did not work as well, because pupils simply looked at the grades.

In addition the temptation for teachers and pupils alike to cram for exams detracted from the long-term benefits of learning how to learn. In other words, the obsession with exams is actually damaging the educational process.

Fortunately, teachers have been less seduced by numerical indicators than the politicians. Many schools and local authorities have looked at Black and William’s research and started to find ways of improving their assessment practices. Those on the ground understand that such a move will ultimately raise standards more than teaching to the test. It seeems perverse to discourage an educational practice that can only benefit pupils.

One of the things I enjoyed most about being an English teacher was watching pupils improve over their two-year, 100-per-cent-coursework GCSE. So many were a testimony to the power of formative assessment, particularly the disaffected boys, who are increasingly falling through the net. Pupils had to make an effort to meet their assignment deadlines, rather than deferring performance to the exam. This, and the instant feedback that coursework offered, enabled them to enter their work for the GCSE after two years, even if for parts of that time they had been absent.

We need official recognition of what a pupil has achieved in school, but ample evidence suggests that, the more we test, the more we restrict that achievement.

Bethan Marshall is a lecturer in education at King’s College, London.