Western Gazette
It has famously been said that there are lies, damn lies and statistics.
I came across an extraordinary proof of this cliché when I visited Marshwood School last week.
Marshwood is an absolutely charming little primary school. It has just about everything you could hope for.
Situated in some of the loveliest countryside in England with spectacular views over the Marshwood Vale, it lies next to the church with which – as a Church of England school – it has close links.
Indeed, the links are so close that the school actually uses the church as a classroom. This is necessary because, despite building an absolutely adorable new classroom amidst the eaves in the attic, the little place is full to overflowing due to its popularity.
The children are bright-eyed and bushy-tailed, enormously polite and enormously friendly. There is a happy buzz; there is a mid-sized turbine; there are uniforms; there is an annual visit to London for several days to open eyes and widen horizons; the headmistress positively gleams with excitement about her school.
One of the things you sense immediately is that, as in so many of our small rural primary schools, the values are right. They are not focusing on league tables. They are focusing on the children. They are trying to do the best, not only academically but also in terms of character and spirit for each child. No doubt that is why the school is so popular.
But even a school like this has to tick all the boxes imposed upon it by the bureaucracy. So, at what is known as Key Stage 2, the achievements of its pupils in basic skills have to be measured.
This is where the statistical quirk comes in .
If you were to compare the schools performance with that of other primary schools at Key Stage 2, you would – as the headmistress pointed out to me – be almost certain to find that they were either right at the top of the league or right at the bottom; and you would see them moving smartly from top to bottom and back again as year succeeds year.
How can this be happening in such an excellent school? Answer: the statistical measures are not designed for small schools like this. They are designed for large schools, in which the average performance is a meaningful item.
Knock the numbers right down to the size of a small rural primary school, and you are measuring the achievements of a handful of pupils.
Result: the performance of just one pupil is enough to change the average result by enough to move from the bottom of the table to the top of the table, or from the top of the table to the bottom of the table.
In practice, it seems that everyone concerned – the teachers and governors, the parents, the inspectors and the local education authority – all recognise this statistical anomaly. So no real harm is done.
But I can well understand the exasperation of this splendid and enthusiastic head teacher at the fact that she has to spend a certain amount of her time explaining that the statistics mean just about nothing.