Understandably, the GCSE English results have attracted the largest share of the headlines overnight and this morning.
The 1.9% drop in grades A* - C in this important subject looks significant. And we are seeing that a minority of schools have seen their results fall beyond expectations, while others have seen improvements.
So let’s consider what is behind these results.
GCSE English saw significant changes this year. Exams were taken at the end of the course. Speaking and listening results were reported separately, and the qualification was rebalanced so that exams counted for more marks than controlled assessments, marked by teachers.
The impact of these changes was taken into account in the awarding process, so that overall students were not disadvantaged, or advantaged. That is why grade boundaries were generally set a few marks lower this year than last.
But changes always affect different schools in different ways, depending on their previous approaches, teaching strengths and the abilities of their students. It was likely that schools that traditionally performed particularly strongly in speaking and listening; those who used a fully modular approach, and those who used resit opportunities to the full would have the biggest transition to make and could expect to see greater variation in their results. Some variation happens every year, that is normal. But evidence tells us that this increases when changes are made to qualifications and the system. And this does appear to be what has happened in some schools.
Indeed, now we have seen the results we have carried out some research into how GCSE English results have varied for schools this year. This has found that those schools that previously chose to use the fully modular approach have seen more variation in results than those who entered students for modules only in the summer. We also looked at schools which had previously made use of re-sitting, and found that those schools had seen greater variation than those that did not. In both cases, the variation tends to be a fall in results. This research can be accessed through our document explaining the GCSE results, along with information on the grade boundaries set for GCSE English this year. We have also looked at other subjects, and that research can be found here. This shows that there was less variation in other subjects where less changes occurred. We will carry out more analysis in the coming weeks, to understand and explain more about how the changes played out in schools.
Separately, looking at the results for the speaking and listening assessments, we see that two out of three students got one of the top two grades (level 4 or 5). That is a consistent picture compared to the results for those assessments last year, and may indicate that free from the accountability pressures, these results have stabilised after years of upward movement.
Something else significant was happening with GCSE English this year, other than the structural changes. The cohort was very different this summer to last. The apparent drop in national results for the summer – the 1.9% at A* - C – reflects those differences. There were significant moves to IGCSE and more students took the qualification in November 2013 instead, the last time that speaking and listening counted towards the grade. That explains why 731,000 took GCSE English last summer but only 516,000 took it this summer – it is not comparing like with like. Elsewhere, different changes in the cohorts are responsible for rises in the pass rates for Maths and Science.
Overall, schools have responded well to the move to exams at the end of the course, and many have seen improved results. The changes mean a more level playing field for all students, as multiple resitting gave some students advantages over others and entry opportunities are now more uniform. We made the decision about speaking and listening because there was no way exam boards could guarantee marking was consistent and fair across all schools. All this mean that people can have greater confidence in these results.