In 30 years of covering education, I have seldom had to write about a drop in exam standards.
The A-level pass rate has gone up year after year, ever since grades were "norm referenced", meaning that a fixed percentage are awarded an A-grade every year. It is a similar story with GCSEs, with the percentage getting five A* to C grades rising year on year.
True, there was a blip in the maths tests for 11-year-olds where the percentage mastering the subject went down by one per cent eight years ago, but it seemed a given that English would just keep on improving – or at least remain at the same standard.
So just how seriously should we take the fact that the percentage of 11-year-olds with a competence in English has fallen for the first time ever?
The answer is that it is not the end of civilisation as we know it. In some ways, it is almost refreshing to know that there is a test where results can go down, as well as up.
Compared to where we were 15 years ago, when more youngsters failed to master the subject (51 per cent) than pass it, there has been considerable improvement.
A study by the widely respected National Foundation for Education Research in the mid-1990s showed there had been no real improvement in standards since the Second World War.
Yesterday's results do, however, underline the fact that there is no room for complacency, and I do have a bone to pick with Labour here.
The party's pledge to give every struggling youngster one-to-one tuition in the "three Rs" will indeed help improve performance among some of our most needy pupils in future. But I have never understood why Labour's simple 1997 election pledge to reduce class sizes for five to seven-year-olds to no more than 30 was never built upon. It seems to me you either believe smaller class sizes will help pupils to learn or they will not.
You could argue that Labour's election promise was of more use to middle-class parents in the shire counties than those whose children were in underperforming inner-city schools. Most of the largest class sizes were in schools run by county councils rather than metropolitan authorities.
Yet research in Tennessee in the US has shown that a major reduction in primary class sizes, to about 15 pupils per class, is needed before it has a substantial impact on performance.
That is why I would be far happier with Labour's performance on education if it had gone on to strive to achieve that as a priority.
There is an argument that the tests – which were ushered into an unsuspecting education world by the then Conservative Education Secretary Kenneth Baker in 1988, in what he modestly called his Great Education Reform Bill – should be abolished, as they take up too much time to give primary school youngsters a balanced curriculum.
There is something wrong with these tests if secondary schools, as happens now, feel they have to re-test youngsters upon arrival because they do not believe the SATs results are an accurate reflection of pupils' abilities.
You could get round that by abolishing the primary league tables – which are why so much coaching is needed for the tests – and taking up a suggestion made by the former senior government adviser Sir Cyril Taylor last month: that all schools should be obliged by law to post information about themselves, including test results, in October each year to give parents the information they need to choose a school for their child.
Alternatively, the idea floated by Michael Gove, the Conservatives' schools spokesman, that the tests should be transferred to the start of secondary schooling, could be explored. (Apparently, it would be possible to track the results back to the child's primary school without too much difficulty). This, it is argued, would reduce the pressure to teach to the tests so schools get a high ranking in the tables.
But it is important that the testing of a child's ability after they complete primary school is retained, both as a means of measuring the school's achievement and a record of what each pupil can achieve as they move on to secondary school.Reuse content