Further contributions to the politics of exams from the Guardian Comment is Free thread here
1. Even Gove now makes quite clear that the old O-levels (which I sat) were for a minority only. People seem to have forgotten that in the 1950s, most students left school at 14 or 15 with no qualifications to their name. The O-level was an exam devised by universities in order to select from those people already selected by the 11 plus, a cohort who would be probably OK for university, medical school, training college and some high-level technical training.
For better or worse, GCSEs were devised as a test for all. The reason why some of them have up until now incorporated course work was because teachers, examiners and society deemed it that this was one way for students to show something that they could do over a period of time ie resembling real life conditions outside. In truth, there is absolutely nothing wrong with this as a process. The problem arises when you try to make this kind of 'valid' work, fit into having a statistically 'reliable' outcome. The Gove interpretation is to say that because it's not 'reliable' , it's not 'valid'. That's upside down thinking. It's indeed very valid because it encourages all kinds of worthwhile work from teachers and students. When 16-plus national exams are abolished, we can start thinking again about making work 'valid' and worrying less about whether they fit these pre-designated norms, so beloved of the exam cultists.
2. Even the 'objective' 11-plus varied from region to region, locality to locality on the basis of how many grammar school places that particular locality provided! There was no national standard called 'passing your 11-plus'. And then it has emerged over the last period that most areas decided that 50% of girls and 50% of boys would pass. However, girls were better at doing the 11plus than boys, so...er... .they fiddled the results, ie marked down the borderline girls and marked up the borderline boys.
So, here were three kinds of test - an IQ, a maths and an English test, carefully and lovingly designed by the exam cultists to be reliable, and then the sociology of the testing made it unreliable. Pathetic.
3. The real absurdity of norm referencing emerges when you relate it to real life. In all learning, we are aware of things we don't know (sorry, but Rumsfeld was right!), and then, if we work at it, we become aware that we have learned something. So, we can all give accounts in our life of change, of learning something.
Every story told from every national testing system will tell you that this day-to-day 'common sense' notion of learning is defied and contradicted by the exam system. The exam system ultimately only exists as a justification for itself, for its claim to be 'reliable'. We are hearing hours and hours of complete crap about whether Ofqual tweaked it in the right or wrong direction at this or that juncture. The whole point is that it was in search of a spurious and useless objective ie that it would produce a reliable result.
In fact, the whole point is that their 'reliable' was the prejudicially determined 'reliable'. What people are saying is that either Gove was introducing this prejudice or it was the political climate, or both.
How can there be any doubt that it was at the very least the political climate at work? There is no 'objective' reliability for exams. It's just a bunch of bureaucrats manipulating young people's lives - ultimately for no other purpose than to earn a decent whack as being on an exam board.
How do I know? Because my dad was one and explained the racket to me many times.