GCSE and A level Government/Ofqual 'Sweetner'

[HappySonHappyMum]

Does anyone else feel that the government/Ofqual have released appeal info and results pattern information in advance of the actual results to try and offset the fury that thousands of teenagers and parents are likely to feel on results days when they realise their kids have been shortchanged? I felt really angry for the Scottish kids when I saw how the data supplied to the SQA had been used. I am feeling even more worried now than I did before. This whole year group are going to lose out and there's nothing these kids can do about it .

I can see that the experts must be satisfied with the theory about the downgrading across the whole cohort, but it just doesn't seem to be producing fair or accurate results at an individual level. That is what is worrying me. At an individual level pupils are getting as much as 4 grades lower (a teacher nearly in tears on the radio just now said that). The problem is that these statistical models haven't looked at individual cases and now obstacles and threats are being put in the way of individual appeals.

teachers/ Heads/ Government etc would be crazy to support over inflated grades and teacher assessment only ( as they have in Scotland) and any school that did this was foolish and short sighted even if it had the short term benefit for this years students.

Inevitably the results will drop again next year (and the year after, and the year after that etc etc ) for the school/ teacher and it'll be obvious the grades were inflated in 2020- how else to explain it?
Equally obvious in the schools that were accurate as their grades won't drop as much.

Scotland will have this problem next year when exams are taken again as normal and some schools will inevitably see a massive drop. What are the Heads/ teachers going to say?

Apparently the statistics are:
39.1% downgraded overall
-35.6% were decreased by 1 grade
-3.5% decreased by 2 or more grades

2.2% upgraded overall
-2.14% increased by 1 grade
-0.06% increased by 2 or more grades

Stats were taken from another thread.
I think it is reasonable to suggest that the normal system for moderation of exam results and NEA results has not worked fairly or accurately at an individual level for these non-exam results based on rankings, previous performance etc. It is not surprising because this year was not a normal year.
What worries me particularly is we have been told we can appeal, we have been told we can use a valid mock result taken under exam conditions, but now schools are being told appeals might affect whole cohort and might not support individual appeals - but how can an individual pupil get a fair review of a grade they feel is unfair when the appeals system seems to be preventing access to that valid result.

@neutralintelligence

I can see that the experts must be satisfied with the theory about the downgrading across the whole cohort, but it just doesn't seem to be producing fair or accurate results at an individual level. That is what is worrying me. At an individual level pupils are getting as much as 4 grades lower (a teacher nearly in tears on the radio just now said that). The problem is that these statistical models haven't looked at individual cases and now obstacles and threats are being put in the way of individual appeals.

surely those getting 4 grades lower must be a teeny tiny number if 2 or more grades lower affects only 3.5 % of the 39% of grades that are lowered - assuming most of that 3.5% are made up of 2 or 3 grades lower. 4 grades lower must be a really tiny percentage and could well be a mistake - as happens every year. So worth appealing those ones I would think but they must be an anomaly not the norm.

I must say the moderation seems to have worked exactly as expected for my subject/school and individual results look very reasonable.

There’s enormous variation in CAGs and Mocks from department to department though. It would be a travesty to allow either of these in appeals.

Hmmm

www.tes.com/news/A-levels-private-schools-get-far-larger-aa-rise

You mentioned NEAS neutral. there is nothing an excellent NEA can now do to form the basis of an appeal which is simply not fair.

Yes, Piggywaspushed, my DS spend months on an NEA which had to be finished in February. Since in February the NEA would form part of his actual marks, he prioritised it over a second round of mocks that wouldn't. Now the NEA (grade 9, apparently A'level grade) won't count, but the mocks will. Upside down, rules of the game changed after the game was finished.

Onions, some type of alternative grade has to be available for an appeal. Normally an appeal would be remarking the paper, but this year there is no paper. So the CAG is the mark the teacher would give and the mock is the mark the pupil actually achieved in a valid exam. Either could and should be used in an appeal. Especially since appeals will need to be very very quick. Sixth form places are normally decided on results day. How will my DS know which sixth form will take him if appeals take weeks?

I think there should have been more of a fight by exam boards over NEAs . But they weren't asked about this 'using the mocks as evidence ' thing and using NEAs for resits is impractical.

NEAs sound like a waste of many months of hard work, staying late for hours after school, going in during the holidays and half term.

If an exam mark can be moderated for an exam that never took place, surely NEA moderation would have been possible. And should be used for resits.

I thought NEAs could be part of the evidence on which the predictions were made?

Yes, so they would have informed the school prediction,

But if that is then graded down, it is unlikely the mocks will help (with a borderline B/A student they often might have a mock of B but coursework at A or A*), and the resit exams have no marked NEA element.

The DfE has forgotten about NEAs in their appeals process.

The issue is NEAs are not like the coursework I did in early days of GCSE. For the DS it was one huge piece of practical work and a huge accompanying document. The scale of it meant that in order to complete it, time had to be taken away from mocks, simply because NEA was part of the actual final mark but mocks were not.
In that respect, some pupils have been cheated because the rules they were following were changed after lockdown and no-one can go back in time and not bother with their NEA and cram for their mocks instead.

Nevertheless, I eagerly await an update from gov/ofqual about how my DS will get his potentially down-moderated result replaced by his mock result, as promised yesterday by the government.
Who's in charge here, government or ofqual?

neuralintelligence, ofqual are a government organisation so most likely the government have told them to allow appeals on the basis of valid mock exams but left the detail (eg how to determine what counts as “valid”) for ofqual to decide.

I think they’ve said they’ll outline the process next week.

OK, I have tried to think why the results do work when the entire cohort of year 13s is taken into account, but are not accurate or fair at an individual level.
What I can see is that: each year many pupils don't perform as well in an exam as the teacher prediction. That may be due to lack of revision, panic, illness like norovirus or a cold, girlfriend/boyfriend issues, going out drinking with those who finished exams already etc. But in those cases, there is an actual exam, the exam board can see the lowered grade, and all grades will be moderated anyway.
This year, there is no paper. No one know who the pupils are who would have underperformed in the exam. The system used here has meant that the pupils whose marks are lower than teacher predictions are automatically those at grade boundaries or at schools that didn't perform as well in previous years. Those consequences have produced a set of statistical assumptions that are incorrect and unfair. So the current moderation of ranking order results does not work at an individual level.

@TheGodmother

My son didn't do well In the SATs I didn't encourage him to. Pointless exercise for a dyslexic child forced to do spelling tests everyday and getting 0. He was upset and I was upset. Weekly meeting with the headmaster. Ended up telling my son not to worry about them and I have no idea of his results because I didn't fucking care!

His mocks' results were atrocious and gave him the big shake up he needed. He studied for 3 hours every day, went to every after school and lunchtime additional studies. I took half term off and helped him study 5 hours a day. He had a maths tutor and an English tutor. He would have definitely got at least 4s maybe even 5s

But all that work and money for tutors will be a complete waste of time. Fuck knows what shit grades he's going to get!

I'm not a teacher but I know 99.9% are dedicated and sensible enough to not over egg students' grades. I thinks it's disgusting that teachers were not trusted to predict the results.

That would definitely benefited my son as they saw the hours he was putting in.

I feel sick about the results this month. This government has let a generation of schools kids down.

40% of grades were pulled down so certainly not 99% of teachers could be trusted with their judgement, more like 80% at best.

I see a lot of criticism, mostly deservedly so but no one offered a workable alternative. A lot of teachers were up in arm about not trusting the professional judgement but 40% is a huge number . Results are still up on previous years as well. It is a total mess.

The system used is not accurate. Its method does not take account of the fact that in a given normal exam year many pupils do not meet the teacher predicted grade. But the reasons for that are generally random and cannot be put in an algorithm. It works in a normal exam year because there is an actual exam paper to mark.
The approach cannot work without an exam paper because pupils ranking order cannot take account of the random factors that would have led to a grade being lower or higher than the teacher prediction. It is not as simple as being a borderline grade or the school having performer better or worse as a whole the year before.
It is not a case of finding an algorithm that is better.
I think it is fairer to accept that this year's grades will be higher due to factors that cannot be resolved in a mass moderation model and still be fair and accurate. The priority is not to make sure that grades are fair to the pupils who took exams last year. The priority should be to make sure that the grades are fair to the pupils who could not take exams this year - those are the people most affected.
CAG or mock result is the only answer.

If a school is thought to have inflated grades by 30%, for example, the algorithm cannot know which particular individual pupils would not have got their predicted grade. It cannot possibly know if that is the pupil on the borderline between grades or a pupil at the top of a grade. There is no way of knowing without an exam paper which pupils would not have got the predicted grade in an exam.
Yet, the algorithm has lowered the grades of individual pupils without needing to provide any evidence that the particular individual pupil in question would not have achieved that grade. That cannot be ethical. Can it even be legal?
There is no way of identifying which individual pupils would not have achieved the CAG. No way at all without an exam paper to mark and review. So mass moderation and this algorithm will never work fairly or accurately at an individual level.
The only option is to revert to CAG or mock, and accept that this is a year group whose result can only reflect the best situation so overall the cohort grades are higher. Anything else takes away grades from pupils without any evidence.

But that would be equally unfair because some schools were clearly more optimistic in their assessments than others.

There is no way of knowing without an exam paper which pupils would not have got the predicted grade in an exam.

No. The pupil rankings were supposed to do that job, but in reality the students who messed up on the day would not have been the ones at the bottom of the grade, they would have been randomly sprinkled through it. Not knowing who they were means that the lower students who would have passed, get unfairly put down.

But then you have the issue of massive grade inflation.

There are no easy answers.

Scotland are happy to have a year with some grade inflation to ensure that each individual pupil gets a fair grade.

It really is not the end of the world for one year to have higher grades overall. But it is really very damaging, harmful and unfair for each individual pupil who has had a grade lowered by one or two grades, sometimes more, without any evidence available that they as an individual would not have achieved that grade in an exam.

@neutralintelligence

It really is not the end of the world for one year to have higher grades overall. But it is really very damaging, harmful and unfair for each individual pupil who has had a grade lowered by one or two grades, sometimes more, without any evidence available that they as an individual would not have achieved that grade in an exam.

^ This ^

If there are centres whose grades are so out of line with previous years, they should be moderated individually and provide the evidence for their assessments. Cheaper and less time consuming than resitting exams and FAR more reliable than mock exams, which for a variety of reasons, are probabyl the LEAST reliable method of assessing students' final exam performance.