School performance tables are a bad way to measure improvement

By Jim Sweetman / @jimbo9848

Labour got saddled in 1997 with using GCSE passes and SAT scores as measures of school performance in annual performance tables in order to appease the standards brigade and OFSTED’s Chris Woodhead. It is time those metrics were dumped.

The publication of school performance tables is politically a no-win event. If there is improvement then it is not enough and if there is a decline – as registered in the last set of SAT results, for example – then it is open season. So, why is the percentage of pupils with five or more good GCSE grades a really bad way to see how schools are doing?

First of all, GCSE grades are not comparable. They should be but they aren’t. Around 60% of children entered for GCSE English Literature get a Grade C or better but for French or any other modern language the figure is closer to 40%. It doesn’t make sense but, if a school needs to be seen to do better, getting rid of modern languages is a good place to start.

To do better still, schools can manipulate the numbers. There is a course in ICT which is worth two GCSE grades so start that up and you can get two grades for the same amount of teaching. You have to teach religious education in a maintained school but you can do a short course in RE and a short course in Citizenship within this teaching time. Each is worth half a GCSE so that’s another way to get your five passes. PE and community languages are another easy option. Taking a GCSE in your own language should get pupils results – but it’s a short-term way to address outcomes, rather than the issues, in any failing school.

Why not just have better teaching and watch standards rise in the subjects you already do? Well, because that is not so easy. Examination standards are set by committees which meet each summer after the marking has been completed. They look at papers and decide where the grade boundaries should go. They are supposed to do this on quality but in practice they often go by percentages. If they recommend a result which will increase or decrease pupils’ percentagages in a grade by more than around 1% they have to explain why in triplicate to the chief executive.

In other words, slow improvement is built into the system but the return wanted by government does not get shown. It always looks as if massive increases in funding are leading to modest improvements and that leads to the level of investment being questioned. Because schools are subject to the same measures as government, they have as great an interest in manipulating the curriculum to boost the output measure. The result is bad for education, bad for pupils and bad for government.

What does bad mean? It means schools spend more on examination fees than on library books and the emphasis on grades above C ignores the achievements of the other 50%, around half of whom still get no recognition of their efforts – under 20% of children get no GCSE grades despite being in those lessons for two years and in school for eleven. Performance tables also set schools up in phony competition with one another when they are meant to be working together to provide new diplomas. And, they distract from the real improvement in schools.

Teaching is better than it was, there is more support for teachers and the environment has been transformed with modernised, safer and brighter learning spaces and new buildings. Of course education has to be accountable but regular objective sampling could the job better. For less money and without all the negative side effects that these tables create we could make schools accountable without burdening them with bureaucracy and yardsticks that are not only inaccurate but also measure the wrong things.