I suppose it is possible to plummet from a centre of excellence to a worst-performing school in three years, but you would have to be pretty determined to do it - good teachers tend to get in the way.
And yet it is true the number of pupils gaining five A-Cs has shown a sharp decline from 67 per cent in 1994 to 43 per cent in 1997. Does this signal a collapse in academic standards or the disciplined ethos of the school? Or are there other possible causes?
There are no visible changes to the social composition. No major new housing estates, no sudden migration to independent schools or other local comprehensives, no large factory closure or opening.
It seems the change is simply due to variation in the natural ability of pupils composing a year group. Since we started to relate GCSE success to Cognitive Ability Test (CAT) scores, we have found the link to be remarkably strong. In 1994, these tests showed that nearly 70 per cent of our pupils were of average ability or above. That year we achieved 67 per cent with five grades A-C and were seen as one of the best schools in the country. This year, only 45 per cent of pupils were of average ability or better and we achieved 43 per cent grades A-C. The link with ability test scores has been equally strong in other years (see figure 1).
Disappointingly, there is no sign of consistent improvement and yet over the past 10 years (let alone five) staff have worked hard at improvement strategies - curriculum development, appraisal, Investors in People, homework clubs, extra curriculum activities, mentoring borderline pupils, records of achievement, careers action plans, etc. A cynic might say if the college is to improve, the key task is to improve the intellectual quality of its intake.
If I were tempted to pursue such a policy, I would never get away with it - thank goodness. I feel confident I can be committed to every child and still command the support and appreciation of my local community, whatever the league tables suggest.
The new requirement to set targets poses interesting dilemmas.
If I had been required to set them for each year group over the past five years, I wonder how I would have explained a declining series of targets. The reality is that
the intellectual potential of different year groups in the college continues to vary but not to decline.
I can be much more optimistic
(but not complacent) about next year's results.
But if I adopt an approach based on the experience of the past five years,I will be setting targets that go up and down. How do I justify such apparently strange variations to parents and pupils? Do I say: "Look, you are not such a good year group as last year's Year 11, so I have set you a lower target?" Year groups are all too easily labelled good or bad without the head leading the process.
And if these targets are met, they will suggest wild swings in "im- provement" on the Government's new indicator. It clearly measures nothing worthwhile at all.
We have to move away from comparisons of performancebased on five A-Cs. It is the worst possible indicator. We must move rapidly to value-added measures - however inadequate our basis might be. League tables based on five A-Cs must be consigned to the dustbinbefore they do even more damage to our educational system. (A more rational approach using the Ringmer exam and cognitive ability test results, with an im- provement target of, say, 2 per cent a year, is suggested in figure 2.)
A similar but much cruder analysis will be possible using national curriculum assessments in five years time when pupils with reliable end-of-primary national test results reach GCSE. The only problem is the vast majority of pupils are squeezed into levels 3 and 4 on transfer and so there is little discrimination in measures of prior attainments.
Such targets cover the whole ability range and can be applied equally to all schools. They enable a school's performance with pupils of all abilities to be assessed and compared, irrespective of the number of pupils in each cohort. One word of warning only - the numbers in each cohort will be relatively small and likely therefore to be unduly affected by one or two children who significantly under or over-perform.
Perhaps the best way forward for a Government committed to inclusive schools is a national programme of cognitive ability testing at 9 and 14 as a basis for measuring school effectiveness.
John Wakely is principal of Ringmer Community College