Friday, December 05, 2003

"Best performing" schools

The opening line of this article from the Sunday Times is untrue or at least unproven. "A QUARTER of Ireland's 30 best-performing secondary schools are Irish-speaking, a study by the Sunday Times has found".

Their study has found no such thing. This study merely found what percentage of students from 400 schools go on to third level education.

I have no problem with league tables or using some other method to evaluate schools' performances. In fact, I think such instruments should be developed, used and the results published. How else can parents assess how their child's school is doing?

But, the Sunday Times measurement does not take into account the standard of the children who have entered each school. There is no attempt to standardize the input, so how can we compare the output?

If a survey shows that Tropicana orange juice is better than Avonmore orange juice, does that prove that Tropicana's factories "perform better" than Avonmore's? No, of course not. It's entirely possible that Tropicana has better oranges going into their juice-making factories.

The same goes for schools.

By all means, let's have school league tables. In fact, let's have schools evaluated and compared on a whole range of factors. Test all kids at 12 and again at 15 and 18. Compare the results. Compare the results for children of different social and economic backgrounds. Perhaps some schools are better at educating middle class children and others better at educating working class children. Some schools may have very few leaving cert students, but they may be outstanding in helping the severely challenged in acquiring the language and mathematics skills needed to get by in life.

But, first, let's stop abusing statistics, mathematics and the English language in the manner that this article does.