Can some one explain to me what the situation with the A- levels is?

[AnaadiNitya]

I’ve tried to have a look at other threads but it’s not very clear ( to me)

Are the exams results being downgraded due to what the school normally achieved? What the lock down anything to do with it?
If a pupil actually achieved a A would it actually be down graded anyway?

Sorry I know I sound dense on the matter.

The big problems have come from classes size 6-15. Because they are so small, the prior years attainment can have a big impact on the allocated grade distribution.

Although for groups of this size, Ofqual did at least use the CAGs alongside the statistical modelling. So my students (a cohort of 13) all received their CAG. My understanding was it was the bigger cohorts that were hit hardest as the CAGs were not used at all to inform the grades given.

I find it very confusing. Even having read most of the links.

Why did the pp who has the one of the biggest 6th forms in the country, get affected so very badly?

@TheOnlyLivingBoyInNewCross

Ofqual used a sliding scale of the 'influence' of the CAGs depending on the group size. However this translated into them largely being ignored. The variations in previous years had a bigger impact. For example a U grade awarded in 2017 because a students missed the exam could mean a U this year in a class of 10 where they were predicted A-C.

Large cohorts tend to be more consistent year on year. So even if their CAGs were ignored, apart from the bonkers U grades, the rest of the distributions tend to be fairer.

The CAGs weren’t used at all for larger schools? Really? So wasting teachers time and ignoring teens abilities? Wtf!!

@Oblomov20 They have been 'affected' in line with national average.

One of the issues is value added from previous years has been largely ignored (although parts of the model do effectively include it). Another issue is the prior attainment of cohorts as no one quite knows how that has been modelled.

One of the biggest issues this year is the perceived unfairness. If you sat an exam and got one grade lower than your prediction, you don't tend to put it out there. This year because your grade was decided by someone else it feels (and is) more unfair. The nationwide grade distributions are pretty much the same. It's easier to stomach messing up an exam yourself over someone else dictating that from the pool of predicted C grades, you'll be the one getting a D.

My understanding was it was the bigger cohorts that were hit hardest as the CAGs were not used at all to inform the grades given

It makes sense it was bigger cohorts as state have been hit harder than independent.

That is pretty outrageous and should not be the outcome. Downgrading should be similar across the board.

@MarshaBradyo

The model tried to standardise between schools. There is no way for ofqual to know if rank 1 person graded A in school A is above or below rank 1 person graded A in school B. Their modelling tried to work this out.

Independents have benefitted from their small classes meaning CAGs were awarded (in most cases). This had a perverse effect of 'taking' some of the allocated high grades away from state schools to keep the distributions similar year on year.

I think the algorithm isn't actually that bad. The issue is the lack of appeals process.

Thank you for the explanations!

Must be absolutely devastating for the pupils who were down graded harshly.

Does any one know why such a model exists or was it quickly put together because of COVID situations?

Quickly put together due to covid. I say quickly, in fairness it probably took a few weeks. There has been some backlash as the Royal Statistics Society offered to check it but ofqual wanted them to sign a 5 year non disclosure agreement. They therefore didn't check it.

Actually having read the OFqual guidance to schools I don't think this is mainly the governments fault. I've had to read the guidance given what has happened to our DS!! Teachers - (for once) were asked their professional opinion. The guidance was CLEAR provide a GRade and a ranking of pupils within that grade. The guidance said to, and I quote
"Use any evidence you have over the entire period of study" The schools did NOT have to submit evidence they merely had to sign a declaration that the grade and ranking they submitted was a true reflection of what they expected to happen had those students say the exam. The biggest problem (from what I can ascertain in our DSs school) is the very narrow and rigid way his school (and probably a difference in the way other schools both favourably and unfavourably used that guidance) has interpreted "any evidence" His school were than also hit with some level of moderation (this is where possibly the government IS at fault) due to the fact his year is only the second year for the school taking A levels and last years year group underperformed. (I know this because our eldest DD was in that year and personally know on of the governors who also has a child in last years and this years A level cohort). However, the fact that our Ds has 2 grades that REALLY don't reflect on him realistically is basically his schools fault not the governments. (And no - neither of those grades should have been an A before anyone wants to say we are just moaning because he didn't get given As!!!)

Thanks for the insight Herc, admittedly this is all new to me. The below seems to be the culprit and a big flaw? If the overall result stacks against state, in favour of independent.

Independents have benefitted from their small classes meaning CAGs were awarded (in most cases). This had a perverse effect of 'taking' some of the allocated high grades away from state schools to keep the distributions similar year on year.

Could someone clarify... cohorts of less than 5 were awarded CAG but what about cohorts between 6 and 15? This is not clear to me?

Also where is the evidence that independent schools were less affected than schools in lower income brackets? I have read this but not seen the numbers to reflect this.

@Hercwasonaroll

Quickly put together due to covid. I say quickly, in fairness it probably took a few weeks. There has been some backlash as the Royal Statistics Society offered to check it but ofqual wanted them to sign a 5 year non disclosure agreement. They therefore didn't check it.

Thank you - and that’s shocking 😮

@ChateauMargaux A combination was used, see the dark line on this graph.

Here's the indie evidence

Yep that's one flaw. In reality there aren't that many indies and even fewer students at them so it probably doesn't have a huge effect.

The flaw is the lack of appeals process and not checking some reductions manually. Eg manually check every U, or even only award a U when the centre has. Also further maths v maths should have been checked as we have students and with an A in FM and C in maths.

It seems that the alithrogram they applied has caused downgrades in schools from more deprived areas. There is a school local to me where a head is campaigning on behalf of a girl from working class background who has got into medical school. She got 3 A in mocks and has been down graded to A C C . Everyone in her chemistry class has been downgraded. She has worked so hard but her individual efforts don't now count because of the use of statistics which is so unfair. And looks at past results for the school and area. She has now lost her place at medical school unless appeal works.

I work in an independent school in an affluent area and we have had some huge downgrades so the idea that private schools have not suffered at all is rubbish. Maybe on average poorer students have but there are still a lot of kids at private schools affected.

I have students downgraded from A or B to D or E. They just haven't looked at the CAGs at all, they have just looked at the rank order and then as we have had students in the past in this subject getting the odd D or E, awarded the lower ranking students in the list D's and E's. No consideration to the fact that this year might be more able or that in fact a few weaker students dropped out so only more able students were left.
It is a completely ridiculous system and so unfair on an individual basis just so they can get a "fair" overall result for grade inflation.

She got 3 A in mocks and has been down graded to A C C .

The head doesn't need to do any campaigning if she got three As in her mocks: the school simply needs to appeal her results. The announcement just before the results were issued added the rider that mock grades would provide a safety net.

Thank you @Hercwasonaroll I get hat the data doesn’t mean anything but I suppose I assumed that a year on year VA increase showed that the school was high performing and not struggling. It’s all so confusing, and I’m a pretty well educated grown up who can’t wrap my head around how this has all been done, what chance have the kids got! I do really appreciate you explaining it all though, an emergency governors meeting has been called for next week so I expect it will be discussed in detail there.

Is it true that none of Eton’s grades have been reduced? I’m seeing it a lot in twitter but wanted to fact check it.

Eton has a large cohort for an independent so I doubt they've had no downgrading.

In my county, almost every school has a sixth form, so in the poorer or more rural towns or villages, those sixth forms have tiny cohorts (particularly doing A levels as opposed to vocational quals). Anecdotally, I believe those schools in deprived areas haven't been as affected as the more middle class schools as they are bigger.

@TheOnlyLivingBoyInNewCross

She got 3 A in mocks and has been down graded to A C C .

The head doesn't need to do any campaigning if she got three As in her mocks: the school simply needs to appeal her results. The announcement just before the results were issued added the rider that mock grades would provide a safety net.

That is another issue with the ‘plan’.

The government released this ‘triple lock guarantee’ the night before but with no real details.

Some schools did low key mocks in the classroom in November-with questions set by class teachers to identify gaps, others did formal ones in exam conditions in the hall with past papers, though some past papers will have been freely available on the internet. Some did mocks in February and marked harshly to encourage children to up their game, other did them in October with encouraging grades to try to build confidence. Some schools planned them for March and then they never happened due to the school closures. Some schools have the results, but no longer have the papers. There is a huge variety of non-moderated information there.

The government have said that they will be releasing information on what counts as a ‘valid’ mock result for the purpose of this triple lock guarantee, but this hasn’t been released yet. If you have missed out on a university place-especially in something like medicine-because of this, and all the places have on the course have now gone, you won’t be going.

I bet the definition of ‘valid’ mock will be something like -it had to take place within January/February, it had to be in exam conditions, it had to be using a past paper which was unavailable on the internet (may be difficult to prove) you have to still have the papers available etc etc.

The media have portrayed it as kids being able to pick the best result they’ve had, but it’s not going to be as simple as that.

Due to lockdown, no exams were taken.

Instead teachers were asked to provide centre assessed grades (CAG) and a rank within each grade. Their very best guess as to what a student would have achieved had they taken the exam. To do this they used all data available to them and knowledge of the students. This would have been checked by head of department and SLT before finalising CAG and rank within each grade for each subject. It was a mammoth and very difficult process.

For every entered student, the school/college sent a CAG and rank to the exam board. Who then applied a standardisation to it to give a final grade.

One of the biggest problems with the standardisation is that small cohorts have been awarded the CAG without having standardisation applied. i.e where less than 15 students take a subject at a school, then the grade the teacher gave has been awarded.

Now, the centres that have the most small cohorts are of course private schools. They will traditionally also achieve the highest grades. So their CAGs will include an awful lot of A/A*grades. These have been awarded without standardisation being applied.

Then the standardisation model is applied to all larger cohorts to ensure that, nationally overall, roughly the same proportion of A/A* are awarded as in previous years.

Trouble is the small cohorts have already taken a large percentage of the top grades without being questioned. This leaves less top grades to be awarded to everyone else.

So small cohort schools (generally private) have been given more top grades than normal while large cohort schools/colleges (generally state) have been left with less top grades than normal.

Overall it may or may not have been fair to downgrade 40% of grades, perhaps some CAGs were over generous, perhaps not. However it is the way the standardisation has been applied (or not) meaning small cohorts do well out of it, where as large cohorts do badly that is grossly unfair and disproportionately affects disadvantaged schools and students.

Our predicted grades were done on a mock they'd just done, controlled past paper (so they couldn't have seen it) completed in exam conditions, marked strictly using official marks gene, then using official grade boundary from that paper. Most of ours were put down one grade. Some were put down two :(