Even if the grade curve distributions are normalised nationally, can't a school, being a small bubble of data in the mix, still raise overall its relative levels of achievement and be justifiably proud of this?
It's interesting to look at the grade boundaries 2024-2025. AQA English language gcse kept the same marks for grades 4 and 5 but lowered boundaries for grades 6-9 ( Edexcel increased boundary for grades 4 and 5 but lowered it for grade 9). ( Similar for maths but the Edexcel changes 2024 to 2025 are massive so it looks like they are saying the papers or marking were off by 20 marks?!). Did they look and decide grade 9 students were not just improving in exam technique but grade 4 and 5 were? Are they saying that the easier questions were too easy and the hard questions too hard? Or that their own markers rewarded exam technique too much at lower grades and ‘actual achievement’ not enough at high grades? I'm happy with any of these being true. But do they do sufficiently thorough moderating for this? Otherwise it looks like insisting on standards more at the lower achievement pass levels (and schools) as David suggests
How? There’s no pegging of results at the individual or school level. (Except in the algo year, which went catastrophically wrong and was cancelled).
They don’t look at a kid or school and work the grades out from there.
Hence, every year loads of schools jump their results up substantially and lots of kids make extraordinary progress.
It’s difficult for schools with kids who enter with low prior attainment because the kids have further to go but comparable outcomes doesn’t make it specifically harder for that school.
Totally agree: Ofqual doesn’t set grades at the level of individual schools or pupils (except, as you say, during the algorithm debacle). Grade boundaries are set nationally, based on cohort-level predictions. No argument there.
But I think we’re talking past each other slightly. The comparable outcomes model doesn’t deliberately make it harder for schools with low prior attainment but in practice, it does create disproportionately higher hurdles.
If your cohort comes in with low KS2 scores, then statistically, they’re expected to achieve fewer high GCSE grades. So even if your teaching is brilliant, and your students make extraordinary progress, you’re still being compared to peers with similar low prior attainment. And since the national model assumes only a small proportion of that group will achieve top grades, your pupils have to massively outperform expectations to break through.
So yes, nothing’s pegged at school level. But the system still exerts pressure through prior attainment, and that plays out unevenly.
Schools with high-attaining intakes benefit from generous expectations — they need only maintain progress to hit the top. Schools with weaker intakes must exceed progress norms to do the same. It’s not a conspiracy. It’s just how distributions work.
So the issue isn’t that comparable outcomes deliberately “cap” results. It’s that they create a statistical gravity that makes equity harder to achieve, especially for schools at the wrong end of the intake curve.
Even if the grade curve distributions are normalised nationally, can't a school, being a small bubble of data in the mix, still raise overall its relative levels of achievement and be justifiably proud of this?
Yes, of course. It’s just especially difficult for schools for low priority attainment
It's interesting to look at the grade boundaries 2024-2025. AQA English language gcse kept the same marks for grades 4 and 5 but lowered boundaries for grades 6-9 ( Edexcel increased boundary for grades 4 and 5 but lowered it for grade 9). ( Similar for maths but the Edexcel changes 2024 to 2025 are massive so it looks like they are saying the papers or marking were off by 20 marks?!). Did they look and decide grade 9 students were not just improving in exam technique but grade 4 and 5 were? Are they saying that the easier questions were too easy and the hard questions too hard? Or that their own markers rewarded exam technique too much at lower grades and ‘actual achievement’ not enough at high grades? I'm happy with any of these being true. But do they do sufficiently thorough moderating for this? Otherwise it looks like insisting on standards more at the lower achievement pass levels (and schools) as David suggests
Ah I see Dominic Salles posted last year's reference test results which explains the different standards applied to lower grades
How? There’s no pegging of results at the individual or school level. (Except in the algo year, which went catastrophically wrong and was cancelled).
They don’t look at a kid or school and work the grades out from there.
Hence, every year loads of schools jump their results up substantially and lots of kids make extraordinary progress.
It’s difficult for schools with kids who enter with low prior attainment because the kids have further to go but comparable outcomes doesn’t make it specifically harder for that school.
Totally agree: Ofqual doesn’t set grades at the level of individual schools or pupils (except, as you say, during the algorithm debacle). Grade boundaries are set nationally, based on cohort-level predictions. No argument there.
But I think we’re talking past each other slightly. The comparable outcomes model doesn’t deliberately make it harder for schools with low prior attainment but in practice, it does create disproportionately higher hurdles.
If your cohort comes in with low KS2 scores, then statistically, they’re expected to achieve fewer high GCSE grades. So even if your teaching is brilliant, and your students make extraordinary progress, you’re still being compared to peers with similar low prior attainment. And since the national model assumes only a small proportion of that group will achieve top grades, your pupils have to massively outperform expectations to break through.
So yes, nothing’s pegged at school level. But the system still exerts pressure through prior attainment, and that plays out unevenly.
Schools with high-attaining intakes benefit from generous expectations — they need only maintain progress to hit the top. Schools with weaker intakes must exceed progress norms to do the same. It’s not a conspiracy. It’s just how distributions work.
So the issue isn’t that comparable outcomes deliberately “cap” results. It’s that they create a statistical gravity that makes equity harder to achieve, especially for schools at the wrong end of the intake curve.