Ofsted have today produced a report
claiming that, for those children leaving primary school with a level 5 in English and Maths, less achieved A/A* in those subjects than in selective schools. However this is not a fair comparison. They overlook that few grammar schools admit every student in their locality who gains level 5+. Most select a higher achieving minority from within this group.
Indeed for many of the elite grammar schools, some of which take students from the top 2% of the cohort, all students will be achieving 5a or higher (as opposed to the lower 5b or 5c SATs grades.)
We do not have access to the student-level data that Ofsted uses. I have previously analysed the data for the school at which I am a governor, and shown how it outperforms
- for those students who were the best achievers at age 11 - when compared to a selective private school like Wellington College.
One journalist, Christopher Cook of the Financial Times, was given access to student-level data and carried out a thorough analysis of the effect of selection. He combined areas using selective education into an invented region called Selectivia and compared how children there performed compared to the actual English regions
- based on primary school performance, poverty and "other stuff". Selectivia comes in the bottom half of regions, with students achieving almost two grades less than those in London.
South West 0.8
South East 0.8
East of England 0.6
North West 0.3
West Midlands 0.1
East Midlands 0.0
North East -0.3
(The benchmark here is the East Midlands and 1 pt equals 1 GCSE grade in one subject.)
He also analysed the effect on the most disadvantaged children, those eligible for free school meals. For those Children Selectivia did even worst, with only one region (the North East) coming below it. (Note that Chris Cook's analysis looked at all school in these selective areas, not just the grammar schools.)
From Christopher Cook's analysis, the conclusion seems clear: Selective educational systems result in lower grades.