When the chief inspector of schools says this, as he did a couple of weeks ago (you may have forgotten, so quickly do things move on), most people will not have the time to explore what he has said in order to contradict him. It’s a serious charge and — if true — one we should take seriously. He claims that secondary schools have consistently been letting down the brightest students and he quotes some compelling evidence to back up what he is saying. However, he is more than wrong – he is also deliberately contributing to the ongoing narrative about teachers with low expectations and an examination system that is not rigorous enough.
It’s difficult to know where to begin. Wilshaw has taken the Key Stage 2 test data and compared what the highest performing pupils in those tests (at the end of their primary school education) went on to achieve at Key Stage 4 (GCSE). The ‘fact’ that so many failed to achieve their ‘potential’ seems to be compelling evidence that something has gone wrong in the intervening five years. But there are other ways of interpreting this ‘evidence’ and, of course, arguments that undermine the ‘evidence’ itself.
- Adolescence : the bright, enthusiastic 11-year-old may lose that enthusiasm as the hormones kick in although, of course, it is part of the teacher’s job to work with that and use teaching techniques that encourage and motivate. Not sure how easy that’s going to be with Gove’s new curriculum and exams!
- Expectations : Wilshaw’s argument is based on what pupils are expected to achieve. Quite recently, schools learned that students were ‘expected’ to achieve ‘three levels of progress’ which means that a pupil gaining Level 5s at Key Stage 2 (the highest possible at that age) ought to achieve at least Grade Bs at GCSE and, for the very brightest, a grade A or A* is expected. Quite what these ‘expectations’ are based on I’m not sure. Not evidence, certainly. And which of these grades should it be? Again, not clear unless you are privy to the data indicating which pupils gained ‘5a’ (towards the top of the mark range) or ‘5c’ (at the lower end, nearly Level 4) which, as far as I can see, is not easy to obtain on a national scale. If a child does not fulfil this ‘expectation’, the secondary school is deemed to have failed. It is actually becoming clearer that A grades are now firmly expected of Level 5 pupils. [I’m told that there is a world of difference between a 5c and a 5a and the suspicion is, that primary schools are nudging the results so that pupils who should get a 4a are ‘tipped’ into the 5c category – see next paragraph].
- Reliability : but, let’s for the sake of argument accept that a child gaining Level 5 at KS2 ought to go on and achieve Grade A at GCSE: how reliable is that Level 5? The answer is, not very. The Key Stage 2 tests are different from GCSE in that they are sat on one day as opposed to consisting of coursework tasks and tests set on an examination day. We also know that Year 6 teachers report starting the preparation for the tests, to be taken in May, as early as the previous September, with students ‘hot-housed’ for months in how to get the highest possible marks. The other thing I have, sadly, to report, is the many anecdotes I have heard over the years of practices that amount to cheating. Not only are primary schools unused to setting up examinations but teachers and headteachers are so desperate for their school to do well that they go way beyond what would be possible in a secondary school where invigilation is done by paid outsiders and teachers are not even allowed in the examination hall. One Year 11 student was told a few months back what grades she was expected to get. Somewhat aghast (she’s a bright student) she asked how these ‘targets’ had been arrived at. A senior member of staff patiently explained to her about KS 2 SATs, three levels of progress and so forth, whereupon she just laughed and described how, back in Year 6, the headteacher had gone round the exam hall where SATs were in progress, looked down at the pupils’ answers and, where he didn’t agree, grunted, took the pen from the pupil and amended the scripts. It’s hard to believe Wilshaw, Gove and their acolytes are unaware of these stories but what have they done to investigate and tighten up? The same is true further back down the line. Why are Year 6 teachers and heads so desperate to do well? The answer is that THEIR results too will be judged against how far pupils have progressed since the Key Stage 1 SATs at the end of infant school. There the reliability is even more suspect, since the work is not externally marked nor even externally moderated. Teachers attend moderation training and, providing they ‘pass’ (ie ‘moderate’ examples in accordance with the guidelines) the marks they award their pupils will be accepted. The whole edifice is constructed on unreliable premises, but Wilshaw wants secondary schools to stream their pupils on the evidence of highly suspect KS2 data. Most secondary schools DO stream from very early on in Maths and English but gave up using the KS2 results a long time ago. Instead, if they can afford it, they’ll use the NFER Cognitive Ability Tests (CATs) which are thought to give an indication of ability and attainment that is untainted by context or background. These USED to be used as the measure by which GCSE targets were set and judged but, guess what, they weren’t ‘rigorous’ enough so we had to start using the far less reliable SATs.
- Catch 22 : having said all of that, does the evidence indicate that in secondary schools, students at the highest level fail to make those three levels of progress? Well, if it genuinely is three levels, no. The evidence I found on line was for the years 2007 – 2011. At Key Stage 2, the percentage of children gaining a Level 5 in Maths and English was fairly level at around 21%. The percentage gaining a GCSE Grade B or higher in Maths over the same period was a little lower, averaging around 16% and for English it was almost the same, at just over 20%. Although results are ‘steady’ there has been a general agreement amongst experts that GCSE English and Maths have got harder over time so that ‘stagnation’ of results is actually improvement, Of course, these were not the same cohorts of children (eg the 2007 KS2 SAT group would have taken their GCSEs in 2012) but does suggest that vast numbers were not being ‘failed’ as Wilshaw has suggested unless, that is, all those Level 5s were ‘expected’ to achieve an A ort A* at GCSE. In that case, he may have a point: in English, over that 2007-2011 timescale, students achieving A or A* were around 11.5 %, in Maths a little lower again at around 10.8% average. However, if Wilshaw is saying the 21% of pupils who got a Level 5 should have gone on to get As or A*s, can we imagine the commentary about ‘dumbing down’, ‘race to the bottom’ and so on if they had actually managed it? This really is a ‘Catch 22’. Teachers are accused of failing their students by not getting them the highest grades in the sorts of numbers that would have had them accused of having dumbed down and ‘failed’ their students!
If you have managed to follow my arguments thus far, well done! No doubt someone much cleverer than me can delve deeper into the statistics and use them to prove me wrong and Wilshaw right – I fear that is the nature of statistics. In summary, however, what I have tried to show is that
- the system is set up to be so competitive between key stages and their teachers that we cannot, in all conscience, claim the KS1 or KS2 results are reliable;
- even if we could, and the expectation that children will make ‘three levels of progress’ is reasonable, the evidence suggests that, looked at one way, they may not be far off doing that;
- looked at another way, if schools HAD managed to get all their ‘Level 5s’ A or A* grades at GCSE, the government and its ‘fellow travellers’ like Wilshaw, would be crying ‘foul’ and claiming the system was letting down our ‘brightest’ in a different way.
So, by all means, let us use the data generated by what is by now decades of tests and results but let the teachers and schools use that data to guide them as to a child’s potential and, by all means, let us have high — but not unrealistic — expectations of all our children. But let’s NOT have politicians or chief inspectors using the data to make inflammatory and inaccurate remarks about teachers.