Alan Smithers argues that rows about falling standards tell us little about pupils - but lots about exams.
August is now firmly the season of exam results. Each year it seems a ritual is enacted. The exam boards announce record results. Sir Rhodes Boyson trenchantly states that things are not what they were. He is supported by various employers, university staff and leader writers. George Turnbull and the exam boards respond, forcefully asserting that standards have been maintained. Pupils and teachers protest that the controversy is devaluing their achievements. The Secretary of State is careful to be on holiday but, when the dust has settled, she returns to claim that the successes have been brought about by the Government's education reforms.
The ritual seems to have taken such hold because there is some truth on all sides. What appears to have happened without it being fully realised is that the function of end-of-school exams has changed. From being about picking out the high fliers they are now about assessing performance against defined levels of attainment.
It is as if the Olympic Committee had decided that gold medals would no longer be given just to those who came first but to all those who achieved a certain standard. Perhaps a gold would go to all those who broke 10 seconds for the 100 metres, a silver to all those below 10.5 seconds and a medal to all those who qualified to take part. Over time probably fewer world records would be broken but the general level of performance would improve. However previous medal winners would protest that now everyone got a medal they were not worth the metal they were embossed on.
With examinations there is an inevitable tension between differentiation, that is, picking out the best, and chalking up credit. The emphasis used to be very much on the former. The 11-plus sent children to different types of school. Only those who got to grammar school, by and large, were allowed a crack at the ordinary and advanced level examinations. These were high hurdles with a predetermined quota for each grade and, in the case of advanced level, a failure rate set at 30 per cent. In this way, it was possible to identify a small number of exceptionally able students who could be educated to degree level in a short time with few dropouts.
It was about winners but it left the majority out in the cold. In order to free up education to develop the talents of everyone the exams have shifted to attempting to assess performance against specified criteria, giving credit accordingly. This has opened the way for the regularly improving results that we have been seeing which sound as if they should be good news.
That they are not accepted as such by everyone seems to be due to at least three reasons. The first is technical. Whereas time can be measured exactly, judging human performance is difficult. With people, pure criterion referencing is a dream. The subtleties of the outpourings of the mind defy simple description. As examiners look back for exemplars to make sense of the criteria, they will take into account the borderline scripts which have been moved up - so there will be gradual slippage.
The second is perceptual. Employers and universities say that they have not noticed much improvement. Their main complaint is that the young people cannot add or spell. This, however, has little to do with GCSE or A-level but reflects, as the recent Social Market Foundation report underlined, what has happened in primary schools where, in the recent past, for the best of motives but with the worst of outcomes, teaching the essentials gave way to self-expression.
The third is that there continues to be a need to differentiate. Places for the top courses in the top universities still have to be allocated, and although higher education as a whole has grown dramatically, the top places by definition remain scarce and prized. If the A and B grades are so plentiful that they no longer distinguish adequately, then the universities will press for something more.
Sir Ron Dearing picked this up in his report on qualifications for 16- to 19-year-olds when he recommended the revival of special papers and the teaching of some university courses (presumably to be examined) in schools. At GCSE we have already seen the introduction of the starred A, suggesting that the A grade is no longer sufficiently discriminating.
We could resolve the tension between differentiating and benchmarking by plumping for one or the other. We could go back to the old competitive arrangements or we could be clear that GCSEs and A-levels or any successor are to be essentially about acknowledging achievement. If universities and employers wanted to select over and above this, it would be for them to do so. But being British, I expect we will continue to try and muddle through attempting to do both together, with the pendulum swinging first one way and then the other. It is likely that our August ritual still has a lot of life in it.
Alan Smithers is professor of policy research and director of the centre for education and employment research at Brunel University