I don't understand why they even need to gives these kids a single grade. The biggest consumer of grading are schools, colleges and university anyway. Just make more of the underlying data available to them and let the decide for themselves. And as an employer I think it would be useful to have a more complete picture of the student than just a single grade. All this fiasco does is demonstrate how unnecessarily reductive the whole concept is.
University entrance requirements in the UK are pretty much always expressed in terms of sets of grades for A levels (rUK) or Highers (Scotland) - often with additional requirements (e.g. must include maths).
The entire entrance process pretty much uses grades as the first filter - changing that would be a huge amount of work not to mention a lot of people will have conditional offers that are expressed in terms of these grades.
So picking a course at random (Physics at Edinburgh) requires: SQA Highers: AAAA - AAAB or A Levels: AAA - ABB
To re-implement the University application process would indeed be a huge amount of work.
However, the governing principal of Cambridge University, for example, is that all operations are done at a human scale, and I think when you factor in the large numbers of staff whose labour the Universities have access to, it becomes possible.
For a group of applicants for Computer Science at a Cambridge college, for example, you would have a Director of Studies and one or two other staff ranking around 20 applicants for half as many places. (The ratio of applicants to places is much higher prior to interview.) It is a tractable problem to solve, even if for other institutions the process is scaled by 2x/3x (4:1 / 6:1, up to Nottingham’s reputed 13:1) or maybe even more.
An idea we mooted over dinner last night was for these applicant rankings to be fed to UCAS — the University and Colleges Admissions Service — and for UCAS to combine these rankings with the applicant preferences to fill up each course from the top down. No need for grades, just use the desirability metric provided by the institutions.
Grades themselves are a strange beast. By their nature they are different to raw marks. A grade speaks to your ability as part of the entire cohort of examined pupils in that subject. Without the exam, taken nationally under examination conditions, it just doesn’t seem to make sense to refer to grades.
If we called them “exam grades” perhaps the folly of trying to assign them to pupils, who have not actually taken an exam, would be clearer.
(I speak with no real authority, but I am a high school teacher in England.)
What underlying data do you want to make available? There's no consistent data collection across the schools, what there is exists for internal use at a school so there's no requirement for it to be to any particular standard or of any particular frequency throughout the course.
It's also a single grade per subject, so the university /employer will usually have 3-5 grades to work with, as well as the broader picture built up from the personal statement/cover letter and interview.
> And as an employer I think it would be useful to have a more complete picture of the student than just a single grade.
The current system isn't coping well with COVID-19 but let's not throw the baby out with the bathwater here.
The current system is simple, and that means it can be seen to be fair. If studying X at university Y needs grades AAB that's easy to understand and you know it's not biased.
I much prefer that to the American system where every university has an opaque system that purports to look at a complete picture of the student then mysteriously picks white students over Asian students with better grades and can't explain why.
> The current system is simple, and that means it can be seen to be fair. If studying X at university Y needs grades AAB that's easy to understand and you know it's not biased.
If only it were that simple. Universities often also take in to account the outcome of in-person interviews which are inherantly biased.
But as a consumer of grades it is not about fairness. It is a heuristic used to predict future performance. And it is possible for grading to be a bad heuristic whilst also being fair to individual students.
The underlying data is just more grades though, isn't it? How you did on this or that exam, coursework, etc? Minus substantial facts like whether you were ill, your gran died, whether you had tutoring, and so on?
You're right that grades are reductive, but the alterntive isn't more of the same information.
That passes the buck on to the university addmissions teams. Then it's the same story with Oxbridge as the antagonists.
Fundamentally it's a hard problem: the admissions require more information than is available. But...
> All this fiasco does is demonstrate how unnecessarily reductive the whole concept is.
...there never really was enough information. This is an important factor IMO. The system has always been unfair, now it's less fair and in a colder, more obvious , more systematic way that resonates with the current political climate around technology. Politically it's hard for this government to get away with standing by a policy that adversely affects poor students in the same way it would have been hard for a left-wing government to institute the current fiscal policy.
>kids
Nit: almost all of them are now legally adults. At university they can look forward to this twilight age where they are treated as an adult when it's about financial obligations and treated as a child when it's their institutions forcing policy upon them.
The Tory line has always been "Work hard and get ahead." It's an outright lie, because the reality has always been "Go to the right school and have the right parents and get ahead."
But much middle class [1] and some working class Tory support still believes in the social status dream, and this fiasco has directly undermined that narrative.
The party is getting huge heat from its base over this, not just because of their future prospects, but because some of them are now wondering if "Work hard and get ahead" has ever been true for them.
[1] Note for the US - in the UK "middle class" specifically means "Highly educated professionals with significant cultural and political capital" not just "People with a degree making decent money."
It's not a single grade. It's a set of usually 3 or 4 grades in subjects the students selected at (US) grade 10 . The choice of subjects contains a lot of information about the students' objectives and the grades show the performance level. The national normalized exam structure is intended to compensate somewhat for large regional and social inequalities. It does not achieve this perfectly, of course.
That's more or less what one of the Oxford colleges did. They realised all grades were guesses and that the information they had when offers were made had not changed quantifiably. So they let all existing offers stand regardless of what grades students were assigned.
I think it's a shame this was the exception rather than the norm.
The Universities know this and are working with students that missed the required grades to understand what their projected grades, mock results etc were and make a case by case evaluation. It's nowhere near as dire a situation as is being made out.
I watched the news broadcasts when the results were announced. There were TV journalists at schools in deprived areas interviewing students before the grades were released, with the journalists giving dour warnings about the impending problems they expected. When the results were announced the students were leaping about with joy, hugging and celebrating, it was like a world cup win. The journalists didn't seem to know what to say. back in the studio the best the presenter could do was mutter that "I'm sure we'll find some students that were disappointed later".
I've been hearing loads of reports of kids getting marks way off what was predicted, and clearly the algorithm is... wrong. This essay is a good explanation of why. One thing that annoys me a bit is that we as a society have discovered all sorts of interesting thing about statistics and decision theory, causality, etc, but then we don't apply them. Why didn't Ofqual find some people who understood how stats work, and and ask them for advice?
It seems clear that if you just take a distribution from previous years and lay them on top of a ranking, you are creating a lot of problems. And you are hiding your incentives on the penalty function, which is something he mentioned. People have correctly figured out that a good kid at a bad school will not be found, and that the marginally second best kid in a year is going to get marked down.
Whatever you think of exams, and I'm no fan, they do give people that mix of noise and signal that society needs to put to rest the issue of who deserves what. They're not entirely a lottery, not entirely skill or hard work, but if we need a way to distribute X number of future doctors, that's the long established way to do it.
>Why didn't Ofqual find some people who understood how stats work, and and ask them for advice?
A very substantial number of the people who work at Ofqual are statisticians of course, it just seems that they have made some serious mistakes.
I don't understand why they couldn't have published at least the algorithm for peer review and discussion months ago. These issues could have been resolved with plenty of time left before results had to be out.
Another problem with this algorithm is that every year, a certain number of exam takers just fuck it up on an epic scale. They didn't get any sleep, they get a question that completely flummoxes them, whatever. So the distribution of grades, even at a good school with a pretty good intake, will have some very low grades on them. A big school will have a U or so. Now come to this year - noone had the opportunity to mess up their actual exam so some poor probably-C, B-on-a-good-day kid has to be stuffed into the "U" to make the distribution fit. Purely on the basis that the big school they went to usually has at least one. The difference is, if you fuck the exam up, you come out knowing that! So when you get your U, you know how it happened. This year someone has had it assigned to them without actually screwing up the exam which seems rather harsh.
> Why didn't Ofqual find some people who understood how stats work, and and ask them for advice?
Because, fundamentally, the current UK government is not interested in scientific evidence or taking advice from experts in an area (c.f. Michael Gove's famous line "people of this country have had enough of experts")
At least include the full quote, which is far more reasonable in-context:
"I think the people of this country have had enough of experts with organisations with acronyms saying that they know what is best and getting it consistently wrong."
Probably best to contextualise that particular quote. Your use of it here is misleading.
The government is on record as wanting precisely more technically qualified people to enter the civil service, and restructuring the service toward this end. Whether or not the way they intend to go about it is sensible or reasonable is another topic for debate.
> I've been hearing loads of reports of kids getting marks way off what was predicted
Why do we need the second set of testing at all if the predictions exist and if the marking is going to be judged against them to the extent we are trusting the prediction over the marking.
>I've been hearing loads of reports of kids getting marks way off what was predicted, and clearly the algorithm is... wrong.
It's worth noting that the predictions that were initially planned to be used as part of the algorithm would have resulted in 38% higher grades than any other year, ever. It's clear that teacher predictions are ALSO "wrong".
It's useful to read some examples to see how devastating this is to children, and how unfair it is that high achieving young people who have done well all throughout their education have suddenly been downgraded.
This is not merely "ham-fisted normalisation". In many cases, the student's attainment is being completely ignored!
> but I think the bigger issue is
The tricky thing about this scandal is that it is actually about 8 different catastrophes for students all enmeshed together. This makes it hard to know where to start critisising the government.
"Gavin Williamson and Ofqual have apologised to students and their parents, as they announced that all A-level and GCSE results in England will [now] be based on teacher-assessed grades." (https://www.theguardian.com/education/2020/aug/17/a-levels-g...)
People should note that even in non-covid times, UK domestic university admissions is almost entirely based on predicted grades. Predicted grades are much more important than actual grades. The only difference this year is an extra layer of noise in the form of an algorithm. It's a very strange, unfair, socially regressive system (I may be a little bitter).
Unless things have changed in the last 24 years * , offers are based on predicted grades, and requirements are set on the actual grades achieved in order for that offer to transform into a place. Those students who did not achieve the requirements and those universities who still have places to fill then enter the "clearing" process to match up.
Were you predicted worse grades than you achieved, and as a result not offered places where you would have liked?
( * they may have, significantly, it's been a long time)
There are a lot of unconditional offers these days, which means that the university will not look at actual A levels' results at all.
Edit:
"38% of applicants (97,045 applicants) received at least one offer with an unconditional component in 2019, increasing from 34% (87,540 applicants) in 2018, and continuing the year-on-year growth in offers with an unconditional component since 2013." [1]
This year at least one Oxford College has already decided to honour all offers without looking at actual A-Level results , whereby effectively turning all offers into unconditional offers [1]. Granted, this is not a normal year.
Let's be honest, though, if you got an offer from Oxbridge you're very good so they're not exactly taking a risk and it gets them brownie points.
I should pre-disclaim this by saying I actually really enjoyed my time at Exeter and it's an excellent school but at the time I felt I'd been treated very unfairly. I think my case is a good example of how the system is badly designed, and then operates in a way that's worse than the theoretical design.
>Were you predicted worse grades than you achieved, and as a result not offered places where you would have liked?
Yes. Story time but skip to the next quote if it's boring :)
I got 4 A's at AS level back in 2001 (I did my Alevels in 2002, the first year of the AS/A2 split). My subject teachers all agreed I'd get the same at A Level. My School decided that "to avoid over estimating" they would reduce all the predictions by 3 levels. So I was predicted BBB. What the Fuck? My friend was friends with the head of VI form (he did theater with her after school, they got drunk together at after show parties, he was very pretty and charming and she was 50+, super weird and dodgy but whatever). Somehow he was predicted AAB. The only one in the school year of ~90 students over BBB. Some people even cried over this as they wanted to do medicine and that's difficult with AAA predicted...
My friend was called for interview at Imperial, where we both wanted to go and study Physics. He went and was offered AAA. I was not called for interview.
Results came in and I got my AAA. He got BBB. Imperial didn't care and admitted him anyway.
I think I am sort of a perfect example of what's wrong with the current system: I was poor and badly turned out and a bit brutish but ferociously bright. He was sort of the opposite. I got good grades, he was better at getting predictions. He was the one that got the Imperial place.
>Those students who did not achieve the requirements and those universities who still have places to fill then enter the "clearing" process to match up.
I think this is the issue because this rarely happens even though it's meant to. There are 2 sides to this:
* If you miss the required grades, you generally keep your place. Universities routinely make an offer at X and then take you whether you get the grades. I understand why: when you reject a candidate with BBB, you're not guaranteed one with AAA, your going to end up with whatever clearing gives you. And not many AAA candidates are in clearing are they? This becomes a self fulfilling prophecy of sorts.
* If you blow through your predicted grades (As instead of Bs) you have 3 options: 1. Stick with the school you were offered at (I did this). 2. Take a whole year out and re-apply next year with your grades. That's a pretty huge step for an 18 years old to take. 3. Enter clearing, but you know that Imperial, Oxbridge etc don't really do clearing do that?
Back in 2002 there was actually discussion of forcing all uni's to keep X% of places for clearing for exactly this reason. It's also worth noting that these days something like 20% of offers are unconditional.
I think I did more than ok in the end. I had a better time than my friend (he had to live at home in first year and never really got the "experience" etc). But people should be aware of how the system works: you're actual grades make very little difference, predicted grades are KEY to getting into (not just an offer, actually into) a top uni, and the best way to get good grades is only about 50% about ability and hard work compared to looking nice and sucking up to the right people.
Maybe this is a great lesson for how life isn't fair :)
The only good reason I can think of is that the UK had a very tight focus on subjects: at GCSE you study 10+ things, at A(s) level 3-4, at university 1.5. You can't really use GCSE results for a UK degree entry because 80% of those subjects are irrelevant for this degree because (unlike EU\North America) a (say) physics degree has zero credits for English lit or history or even chemistry. A Level results are too late. So that puts everything on AS results (and I think the Tories got rid of the actual AS exams a few years back?).
I am a big "final exams not coursework please" guy. But they'd be better having end of term exams for the 6 terms in A Levels and using the results from the first 5 for most of admissions IMHO.
The civil service? Pretty sure this is Dominic Cummings overriding the civil service. But there's plenty enough wrong in this story to avoid invoking his name.
There are plenty of data available on each students. The issue is how to normalise them in order to create a national ranking.
Mock exams may be useful because, arguably, the results should already be pretty normalised, but I'm not sure if these took place this year.
So what is left is to work with is the grades the students have had throughout the year and previous years and, importantly, also what school they are in.
The latter is heavily charged politically so they are wary of communicating on how that impacts their model.
It seems to me, though, that pupils from top private schools are not complaining too much about having their grades downgraded, so it seems that it did not happen too much for them.
On the whole, students from private schools had significant grade inflation over previous years, often because the predictions were used as-is. This left fewer high grades "available" for state schools (in order to maintain the whole-country numbers).
Readit News
The entire entrance process pretty much uses grades as the first filter - changing that would be a huge amount of work not to mention a lot of people will have conditional offers that are expressed in terms of these grades.
So picking a course at random (Physics at Edinburgh) requires: SQA Highers: AAAA - AAAB or A Levels: AAA - ABB
https://www.ed.ac.uk/studying/undergraduate/degrees/index.ph...
However, the governing principal of Cambridge University, for example, is that all operations are done at a human scale, and I think when you factor in the large numbers of staff whose labour the Universities have access to, it becomes possible.
For a group of applicants for Computer Science at a Cambridge college, for example, you would have a Director of Studies and one or two other staff ranking around 20 applicants for half as many places. (The ratio of applicants to places is much higher prior to interview.) It is a tractable problem to solve, even if for other institutions the process is scaled by 2x/3x (4:1 / 6:1, up to Nottingham’s reputed 13:1) or maybe even more.
An idea we mooted over dinner last night was for these applicant rankings to be fed to UCAS — the University and Colleges Admissions Service — and for UCAS to combine these rankings with the applicant preferences to fill up each course from the top down. No need for grades, just use the desirability metric provided by the institutions.
Grades themselves are a strange beast. By their nature they are different to raw marks. A grade speaks to your ability as part of the entire cohort of examined pupils in that subject. Without the exam, taken nationally under examination conditions, it just doesn’t seem to make sense to refer to grades.
If we called them “exam grades” perhaps the folly of trying to assign them to pupils, who have not actually taken an exam, would be clearer.
(I speak with no real authority, but I am a high school teacher in England.)
It's also a single grade per subject, so the university /employer will usually have 3-5 grades to work with, as well as the broader picture built up from the personal statement/cover letter and interview.
The current system isn't coping well with COVID-19 but let's not throw the baby out with the bathwater here.
The current system is simple, and that means it can be seen to be fair. If studying X at university Y needs grades AAB that's easy to understand and you know it's not biased.
I much prefer that to the American system where every university has an opaque system that purports to look at a complete picture of the student then mysteriously picks white students over Asian students with better grades and can't explain why.
If only it were that simple. Universities often also take in to account the outcome of in-person interviews which are inherantly biased.
You're right that grades are reductive, but the alterntive isn't more of the same information.
Fundamentally it's a hard problem: the admissions require more information than is available. But...
> All this fiasco does is demonstrate how unnecessarily reductive the whole concept is.
...there never really was enough information. This is an important factor IMO. The system has always been unfair, now it's less fair and in a colder, more obvious , more systematic way that resonates with the current political climate around technology. Politically it's hard for this government to get away with standing by a policy that adversely affects poor students in the same way it would have been hard for a left-wing government to institute the current fiscal policy.
>kids
Nit: almost all of them are now legally adults. At university they can look forward to this twilight age where they are treated as an adult when it's about financial obligations and treated as a child when it's their institutions forcing policy upon them.
But much middle class [1] and some working class Tory support still believes in the social status dream, and this fiasco has directly undermined that narrative.
The party is getting huge heat from its base over this, not just because of their future prospects, but because some of them are now wondering if "Work hard and get ahead" has ever been true for them.
[1] Note for the US - in the UK "middle class" specifically means "Highly educated professionals with significant cultural and political capital" not just "People with a degree making decent money."
I think it's a shame this was the exception rather than the norm.
I watched the news broadcasts when the results were announced. There were TV journalists at schools in deprived areas interviewing students before the grades were released, with the journalists giving dour warnings about the impending problems they expected. When the results were announced the students were leaping about with joy, hugging and celebrating, it was like a world cup win. The journalists didn't seem to know what to say. back in the studio the best the presenter could do was mutter that "I'm sure we'll find some students that were disappointed later".
I'm sure they did.
It seems clear that if you just take a distribution from previous years and lay them on top of a ranking, you are creating a lot of problems. And you are hiding your incentives on the penalty function, which is something he mentioned. People have correctly figured out that a good kid at a bad school will not be found, and that the marginally second best kid in a year is going to get marked down.
Whatever you think of exams, and I'm no fan, they do give people that mix of noise and signal that society needs to put to rest the issue of who deserves what. They're not entirely a lottery, not entirely skill or hard work, but if we need a way to distribute X number of future doctors, that's the long established way to do it.
A very substantial number of the people who work at Ofqual are statisticians of course, it just seems that they have made some serious mistakes.
I don't understand why they couldn't have published at least the algorithm for peer review and discussion months ago. These issues could have been resolved with plenty of time left before results had to be out.
Another problem with this algorithm is that every year, a certain number of exam takers just fuck it up on an epic scale. They didn't get any sleep, they get a question that completely flummoxes them, whatever. So the distribution of grades, even at a good school with a pretty good intake, will have some very low grades on them. A big school will have a U or so. Now come to this year - noone had the opportunity to mess up their actual exam so some poor probably-C, B-on-a-good-day kid has to be stuffed into the "U" to make the distribution fit. Purely on the basis that the big school they went to usually has at least one. The difference is, if you fuck the exam up, you come out knowing that! So when you get your U, you know how it happened. This year someone has had it assigned to them without actually screwing up the exam which seems rather harsh.
They wouldn't sign a five-year NDA:
https://news.sky.com/story/a-levels-exam-regulator-ignored-e...
The nicest interpretation of this is that Ofqual had no idea what the hell they were doing.
Because, fundamentally, the current UK government is not interested in scientific evidence or taking advice from experts in an area (c.f. Michael Gove's famous line "people of this country have had enough of experts")
"I think the people of this country have had enough of experts with organisations with acronyms saying that they know what is best and getting it consistently wrong."
The government is on record as wanting precisely more technically qualified people to enter the civil service, and restructuring the service toward this end. Whether or not the way they intend to go about it is sensible or reasonable is another topic for debate.
Why do we need the second set of testing at all if the predictions exist and if the marking is going to be judged against them to the extent we are trusting the prediction over the marking.
It's worth noting that the predictions that were initially planned to be used as part of the algorithm would have resulted in 38% higher grades than any other year, ever. It's clear that teacher predictions are ALSO "wrong".
https://twitter.com/lewis_goodall/status/1295069110829752323...
And this describes the advantages that private schools got: https://twitter.com/hugorifkind/status/1295292925463724032?s...
I think it should be possible to create ‘virtual classes’ above the threshold by combining smaller like classes together.
Such a model would be fairly ham fisted.
> but I think the bigger issue is
The tricky thing about this scandal is that it is actually about 8 different catastrophes for students all enmeshed together. This makes it hard to know where to start critisising the government.
Were you predicted worse grades than you achieved, and as a result not offered places where you would have liked?
( * they may have, significantly, it's been a long time)
Edit:
"38% of applicants (97,045 applicants) received at least one offer with an unconditional component in 2019, increasing from 34% (87,540 applicants) in 2018, and continuing the year-on-year growth in offers with an unconditional component since 2013." [1]
This year at least one Oxford College has already decided to honour all offers without looking at actual A-Level results , whereby effectively turning all offers into unconditional offers [1]. Granted, this is not a normal year.
Let's be honest, though, if you got an offer from Oxbridge you're very good so they're not exactly taking a risk and it gets them brownie points.
[1] (PDF) https://www.ucas.com/file/250931/download?token=R8Nn7uoI#:~:....
[2] https://www.theguardian.com/education/2020/aug/15/a-levels-r...
Motivation for working hard during 6th year can be difficult if you get an unconditional offer based on results you already have....
>Were you predicted worse grades than you achieved, and as a result not offered places where you would have liked?
Yes. Story time but skip to the next quote if it's boring :)
I got 4 A's at AS level back in 2001 (I did my Alevels in 2002, the first year of the AS/A2 split). My subject teachers all agreed I'd get the same at A Level. My School decided that "to avoid over estimating" they would reduce all the predictions by 3 levels. So I was predicted BBB. What the Fuck? My friend was friends with the head of VI form (he did theater with her after school, they got drunk together at after show parties, he was very pretty and charming and she was 50+, super weird and dodgy but whatever). Somehow he was predicted AAB. The only one in the school year of ~90 students over BBB. Some people even cried over this as they wanted to do medicine and that's difficult with AAA predicted...
My friend was called for interview at Imperial, where we both wanted to go and study Physics. He went and was offered AAA. I was not called for interview.
Results came in and I got my AAA. He got BBB. Imperial didn't care and admitted him anyway.
I think I am sort of a perfect example of what's wrong with the current system: I was poor and badly turned out and a bit brutish but ferociously bright. He was sort of the opposite. I got good grades, he was better at getting predictions. He was the one that got the Imperial place.
>Those students who did not achieve the requirements and those universities who still have places to fill then enter the "clearing" process to match up.
I think this is the issue because this rarely happens even though it's meant to. There are 2 sides to this:
* If you miss the required grades, you generally keep your place. Universities routinely make an offer at X and then take you whether you get the grades. I understand why: when you reject a candidate with BBB, you're not guaranteed one with AAA, your going to end up with whatever clearing gives you. And not many AAA candidates are in clearing are they? This becomes a self fulfilling prophecy of sorts.
* If you blow through your predicted grades (As instead of Bs) you have 3 options: 1. Stick with the school you were offered at (I did this). 2. Take a whole year out and re-apply next year with your grades. That's a pretty huge step for an 18 years old to take. 3. Enter clearing, but you know that Imperial, Oxbridge etc don't really do clearing do that?
Back in 2002 there was actually discussion of forcing all uni's to keep X% of places for clearing for exactly this reason. It's also worth noting that these days something like 20% of offers are unconditional.
I think I did more than ok in the end. I had a better time than my friend (he had to live at home in first year and never really got the "experience" etc). But people should be aware of how the system works: you're actual grades make very little difference, predicted grades are KEY to getting into (not just an offer, actually into) a top uni, and the best way to get good grades is only about 50% about ability and hard work compared to looking nice and sucking up to the right people.
Maybe this is a great lesson for how life isn't fair :)
But using grades pupils have achieved during the previous years(s) seems fine to me and can potentially erase the issue of having a good/bad exam day.
In some countries, either they use exam results and/or they look at report cards for the previous years.
I am a big "final exams not coursework please" guy. But they'd be better having end of term exams for the 6 terms in A Levels and using the results from the first 5 for most of admissions IMHO.
That explains more than anything what the problem is with the British Civil Service. The epitome of "nobody every got fired for buying IBM".
Mock exams may be useful because, arguably, the results should already be pretty normalised, but I'm not sure if these took place this year.
So what is left is to work with is the grades the students have had throughout the year and previous years and, importantly, also what school they are in.
The latter is heavily charged politically so they are wary of communicating on how that impacts their model.
It seems to me, though, that pupils from top private schools are not complaining too much about having their grades downgraded, so it seems that it did not happen too much for them.