Academic selection does not stop at 11. Many of the grammar pupils receiving their GCSE results in Northern Ireland last week required a minimum total score, typically equivalent to six grade Bs, to be allowed to stay on at their current school.
Many also needed a minimum grade in each subject they wished to continue at A-level, again typically a B.
There is little sympathy with the idea that a young person might pursue a subject because they find it interesting, or because it broadens their general education – or that pupils should be encouraged and helped with a particular subject if they have the potential to do better.
Some grammars get away with this by operating a legally distinct sixth form college, with its own entry requirements. Others just blatantly chuck out any children they consider to be underperforming.
However, pupils sacked at 16 can at least start their A-levels elsewhere. A number of grammars in Northern Ireland weed out pupils half-way through sixth form, using AS level results, leaving them with no way to complete their qualifications.
The Department of Education says schools should not set any re-entry requirements between lower and upper sixth but headteachers and boards of governors ignore the guidelines and do what they like – the real grammar ethos.
Read more:
Relief for many GCSE students as results envelopes opened
A-level grades in Northern Ireland fall, but remain higher than pre-pandemic levels
Newton Emerson: Can the Northern Ireland Civil Service be fixed?
Weeding out begins even before GSCEs, with schools steering pupils away from 'difficult' subjects or reducing the number of GCSEs they can take, depending on their test results up to age 14.
This process of continuous selection is the opposite of teaching and makes a mockery of the supposed purpose of the transfer test, which is to place children in an appropriately supportive educational environment.
If it is not possible to identify an academic child at 11 and nurture their potential in a grammar from that point forward, then selection is pointless. If it is possible, yet is not being done, grammars are refusing to do their job.
The suspicion that grammars are continuously selecting children instead of teaching them is supported by new research in England, published this week in the peer-reviewed Educational Review.
It finds no significant overall difference in GSCE grades between areas with academic selection and areas with fully comprehensive systems. Grammars do not improve results, even for the brightest children; they mainly cream off the brightest children to burnish the school's brand.
Of course, similar findings turn up all the time and parents do not care. The appeal of grammars is at least as much about culture and discipline – and social selection, frankly – as about education.
Opponents of selection need to recognise that reality but supporters need to accept that getting into a 'good' school does not ensure a good education, or even a guaranteed place beyond 16. In the worst cases, it merely enters children into a game of middle-class It's a Knockout.
This game is driven by league tables, with some grammars working to the test – some would say fiddling the results – by removing pupils who might lower the school's score.
League tables are compiled by government. They were introduced to the UK in 1988 by landmark Conservative legislation intended to drive up standards. There is some evidence this has worked, including negative evidence – after the Welsh government abolished league tables in 2001, exam performance fell. Researchers at Bristol University blamed this on removing the "name and shame effect".
Stormont's Department of Education, like its Scottish counterpart, places a standard disclaimer on its tables, noting a simple ranking of schools by exam results "does not provide a valid basis for comparing performance... since it takes no account of intakes or any other factors".
Northern Ireland is unique in the UK for its extensive use of selection, effectively in defiance of the abolition of the official transfer test 15 years ago. So there is a strong case for Stormont producing more sophisticated statistics that try to adjust for schools gaming the system.
The media could also play a role. Newspapers have been crucial to publicising and popularising league tables. Some compile their own, although none really stray from a simple ranking.
By adjusting tables for how many pupils a school retains to age 18, grammars would be incentivised to teach struggling children instead of being incentivised to remove them. Simple rankings could be published as well. If selection is worth having, grammars should do well in both sets of figures.
Armed with such information, the education debate in Northern Ireland might look rather different.
