It's not cheating exactly... is it?
the moment of revelation came when I sat in on a training event at a west London hotel, in which a very senior examiner gave teachers (each paying Pounds 200 for his words of wisdom) tips on preparing pupils for French GCSE.
These 18 professionals were effectively told how to cheat their way through coursework and to "script" oral exams, reducing them to a test of pupils' ability to remember pre-prepared answers. They were also advised not to bother teaching much grammar.
The coursework, he said, could be scripted by the teacher to give pupils a "store" of marks to fall back on. The key was to ensure they did better than their target mark but not so well as to arouse the exam board's suspicions.
Teachers, he said, should tell pupils to write down phrases in a special place in their exercise books. Some schools, he said, labelled this place the children's "Burglar Bill swag bag".
"If you do that, you are in with a shout," he said. "Children I like this idea of stealing."
Pupils could be given examples of, for instance, the perfect tense, which they could later retrieve and insert into their coursework.
To hesitant laughter from his audience, he said: "It's not easy getting them to transfer things from one piece of paper to another!"
Another tactic, he suggested, was to be lenient with coursework marking. I witnessed the following discussion.
Teacher: "If you judge them to be a 35, can you I give them 37 or 38?"
Examiner: "That's right, if it's within the (board's) tolerance."
Teacher: "So, be generous?"
Examiner: "Yes, be realistically generous."
The examiner added that teaching French grammar to less-able children, whom he described as "grim", might be a waste of time, as the exam's listening and reading sections were largely vocabulary tests. And pupils could be told what to write for coursework.
For the oral presentation, children could learn a speech to recite back to their teacher and then answer questions that they have been given in advance.
For the oral conversation, he advised spending two years getting pupils to learn the answers to a list of 42 possible questions. The teacher could then choose a few to ask on the big day.
The next day, I attended a seminar in the North West on achieving an A* in GCSE history, led by another senior examiner who has great influence in the design and marking of papers.
The pattern was similar, the examiner observing that the standards required to achieve A* were not that high. Examiners grade papers using a four-level mark scheme. Level 4, the highest, was not required.
He said: "You might, on reflection, rather than getting students to level 4, just get them working at level 3, which is good enough."
A private school teacher I spoke to seemed persuaded.
The examiner also advised that as the exam approached, schools that were worried about results should concentrate on questions requiring analysis of historical sources, since little factual knowledge was needed to do well in this section.
Teachers should also use coursework to improve pupils' performance in the exam hall, by ensuring their assignments closely matched predicted exam questions.
Morally, this may sound dubious. But, as the examiner added: "This is realpolitik, a question of ends justifying means. The grade boundaries are lower than you might think."
This edited extract is from Warwick Mansell's 'Education by Numbers: The Tyranny of Testing', published by Politico's (pound;19.99) next month
EASY RETAKES NET GLOWING RESULTS
Pass rates for one of Britain's toughest A-levels are being enhanced dramatically by allowing pupils to retake easier papers to conceal poor marks in the harder ones.
A Government analysis of A-level physics results from the AQA's 2006 exam shows that two of the AS papers were taken at least twice by 30 per cent of students. In both, a clear majority of candidates improved their marks at the second go. For the second paper, three-quarters did.
This appears to have had a marked effect: some 24 per cent of students achieved an A grade on the harder A2 papers, which few students retake. On the easier AS, the figure was 48 per cent.
And while 20 per cent failed on the A2 papers, at AS the failure rate dropped dramatically to 0.5 per cent. The overall pass rate for the A-level, which is based on adding students' AS and A2 marks, was 95 per cent.
In other subjects the effect was less severe, the analysis of A-level results in five subjects for the Qualifications and Curriculum Authority found.