Can someone explain why grade boundaries change

[Ceci03]

I'm still struggling to understand. Oning from Rep of Ireland where an A is always 85-100 and a B is always 70-84 I'm struggling to understand why the boundaries change for A level. And it seems for GCSE as well? Like how does a student know how they're doing when the boundaries are so fluid. I dont understand. Thanks !

So employers or members of the public or anyone should never compare different cohorts is that what you're saying. Cos I hear comparisons all the time . Like parents or neighbours or kids themselves comparing themselves. Like saying oh X got AAB and Y got BBC so X is cleverer than Y. Like if you're an employer would you not go for the AAB candidate? How can unis have the same entrance requirements every year. There just seems to be a lot of luck involved. I think I must be very thick cos I'm not gettin it

I mean who knows is you're a candidate who did well to get 70% on an hard paper or who cruised and got 80% on an easy paper. Im stupid aren't I

Fairer than just giving set grades no matter how hard or easy the paper is IMO

I’m in Scotland and it’s the same here broadly it’s 50% for a C, 60 for a B, 70 for an A. But the boundaries are tweaked so broadly the same percentages get the same grade each year. Seems better than Ireland to me where a good student could either be penalised by a particularly hard paper or their grade is devalued in an easy paper where lots get the top grade.

So employers or members of the public or anyone should never compare different cohorts is that what you're saying.

No, this is the opposite of what I'm saying. Nobody should compare percentage scores of people who sat different exams, but comparing grades is fine, that's the whole point of doing it this way. The system is designed so that a kid of a certain ability would get the same grade whichever year's exam they took.

(Ignoring the extra complications of covid - 2020, 2021 and 2022 grades can't really be compared to any other year.)

Here’s how it has been done - more information than most people would want.

I tried to summarise the document several times, but couldn’t do it very well. Here’s my best effort.

For A level, grade boundaries for A and E grades are set (p4) for each paper using information from from the current and past year (p6) and statistical information comparing this year with previous ones.

p7 has my favourite bit of detail. Many experienced examiners look at the question papers and answer schemes and marked scripts for the current and previous years.

Once the A and E grade boundaries are set, then the B, C and D boundaries are calculated to be evenly spaced between them. That’s arithmetic, it doesn’t depend on how many people get each mark.
There’s a detailed example on p5.
A simpler one might be A: 200 E:100.
The difference is (200 - 100).
Divide that by 4 to get the gap between boundaries : 25.
So, the grade boundaries are A:200, B:175, C:150, D:125, E:100.

It’s a similar process for other exams, but possibly starting with other grades (p4)

There’s a summary chart on p9.

It’s not graded on a bell curve. I associate that with the USA, not sure why.

The system is designed so that a kid of a certain ability would get the same grade whichever year's exam they took.

This.

Do you not see @Ceci03 ? It was spectacularly stupid of your dd's tutor to take a mark from one year and say "it would have been a B in xx year". Your dd didn't sit the exam last year, with the issues faced by those dc. If she had, yes she would probably have got a lower mark and still the same grade. Of course you can compare grades.

I wonder what you make of the fact that in GCSE maths, you can get a grade 4 on the higher paper with a score of 28% or whatever it is; but on the foundation paper a 4 requires ? 70% (or whatever; not looked up figures). Why? Bc the F paper is easier! But both grades are a 4 and thus can be compared.

@Ceci03 the whole point of the exercise is to standardise the results and to make it possible to compare the grades across different cohorts, fairly.

Each year the exam boards write new exam papers, with different questions, and it's not possible to make them exactly the same level of difficulty as the previous years given that they ask different questions in different ways. Hence the need to standardise and adjust the grade boundaries each year.

Rather than an exam think about something like athletics.

At major competitions such as the Olympics you expect records to be broken, but it doesn't mean eg the person who has jumped the furthest on the long jump gets the record because they take in to account the wind speed.

A tail wind gives you an advantage so you will get the whatever medal is they have won but not the record.

The same if you were running 100 metres, the wind makes a difference.

Imagine if someone were running 100m but they are running uphill, they still run the same distance as some on the flat, or someone on a downhill trajectory but it will be harder and take longer.

Moving the grade boundaries is about fine tuning to give a fair result.

Sometimes a problem is noticed in a paper that makes things harder. Eg (this is a real one) a physics paper ask student to calculate the energy used to boil a kettle.

They specified the size and power of the element but didn't say if the element was in the kettle or on the hob with the kettle sitting on top.

So students were split 50/50 of how they answered the question. If you stuck strictly the the answers given in the marking scheme 50 % of students would have got 0 for that question.

So they instructed those marking to mark both types of answers and I believe there might have been an adjustment because time might have been wasted deciding which to do.

Another thing that happened one year was that a hair had somehow got stuck at some point during printing so a 1 became a -1.

I find the whole concept of different exam boards really odd. Why doesn’t everybody just sit the same paper, then there wouldn’t need to be a cottage industry around it? I assume it’s because someone somewhere is making a lot of money out of it!!

Well not all boards cover the same aspects of a subject, take RE, an RC school is going to pick GCSE RE that looks at RC values, they do have to look at one other faith but they don't need to look at any more than that.

A school in an inner city that is culturally and religiously diverse might pick a more philosophy and ethics syllabus.

Not all schools in the UK teach in English so there are GCSEs in Welsh for native speakers and non native speakers.

The point is that the person who got 70% on the hard paper would get the same grade as the person who got 80% on the easy paper. Because the grade boundaries would move to allow for this. So the employer (or whoever) knows they're both an A standard candidate.

Whereas in the system from your childhood one would get an A and one would get a B. No one would realise one paper was harder than the other.

The whole point of the system is to do exactly what you're saying it doesn't do! That is to make it more consistent when comparing kids from different year groups.

@MathsIsAVitalSkill
There is some sort of curve used though because it's not the same proportion of people who get each grade. 3% get a 9, something around 15% get a 4. So there is a consideration of a normal distribution, like you'd expect wirh any large population.

https://analytics.ofqual.gov.uk/apps/GCSE/9to1/
See here for distribution by subject.

I still feel students are at a disadvantage here as you never quite know what you need to get to achieve the grade you're looking for

Exam boards write detailed mark schemes (in advance of the tests) that show students how they gain marks in an exam. Most schools will mark exams during the year using the mark scheme so students are familiar with the basis of how they are assessed. Has your daughter explained this? If school didn't explain this then perhaps they have been too optimistic in her predictions.

I appreciate you are struggling to understand however there is rigour in the exam system. If your daughter got a D then it was on the basis of how she did in that exam against the marks scheme. You can check the grade boundaries for her exam board and subject as it's all published.

I don't think it was helpful for the teacher to say "she would have got a B in another year" because it isn't likely to be the case. Look up the grade distribution between this year and last for the subject.

As an aside I don't think you can compare results equally unless you know 2 students doing the same subjects, in the same year, which rarely happens. Was this GCSE or A level? Both are usually stages where you go to do something afterwards so she has the same chance as everyone who gained a D in that year for that subject.

You have to ignore the 2020 and 2021 grade boundaries which were resits for the Covid cohort students and the exams taken in 2022 where there were extra adjustments for students that won't be available in future years.
So the grade boundaries you need to take notice of are the ones for 2019 and 2023 which are comparable.
In normal years, pre-covid, grade boundaries only change by a very small amount each year.

I appreciate you are struggling to understand however there is rigour in the exam system. If your daughter got a D then it was on the basis of how she did in that exam against the marks scheme. You can check the grade boundaries for her exam board and subject as it's all published.

I don't think it was helpful for the teacher to say "she would have got a B in another year" because it isn't likely to be the case. Look up the grade distribution between this year and last for the subject.

So the thing is, if dd was marked to last years mark scheme she would
Have gotten a higher grade . So that is what she was aiming for does that make sense. The last papers and mocks she did were marked to that mark scheme. So it seems to leave students on "shifting sands" as she thought she did well on the paper and she did according to what she'd been practicing on past papers. Does anyone get what I'm saying

Yes. Hence why they change the grade boundaries to accommodate? So you can't cruise

Your DD would not have got a B the year before as that was a different paper, with different mark schemes and different grade boundaries. And different circumstances.

No student should ever sit an exam and go "I need 50% to get a B". They go in and try their best and get the grade that reflects their best work

Maybe that's where her teachers and grinds went wrong then. She spent a lot of time doing all the past papers from 2020, 21, and 22. None of them mentioned anything about marking to a 2019 scheme. I dunno. Obviously thousands of students did really well this year and got what they wanted. Dd just feels like she is "stupid"

Your Dd didn't sit last years paper though did she? So she can't be marked against the mark scheme for a paper she didn't sit under exam conditions.

No student should ever sit an exam and go "I need 50% to get a B". They go in and try their best and get the grade that reflects their best work

For real? Even at uni we were always told to "Aim for xyz" and then you will be sure to get the grade you want?

Your Dd didn't sit last years paper though did she? So she can't be marked against the mark scheme for a paper she didn't sit under exam conditions.*

Where does that leave her then. Seems like nobody could have predicted what she would get in that case. She did past papers and mocks and got all As . She was studying to that standard. But it wasn't good enough for 2023 mark scheme. Does anyone understand what I'm trying to say?

I don’t think you are stupid op but I do think you must be wilfully misunderstanding when so many people have clearly explained it to you.

Working to just try and get a percentage is just daft as there’s always a risk it will be a mark or two out.

Exam boards both Scotland and England work this way so it’s hardly unusual and seems to have served everyone fine for decades?

Some people just have off days in exams, their nerves get the better of them, or their teachers’ predictions were unrealistic:

I think your mistake has been looking at past papers and saying oh that will be xx grade in an exam when you really can’t take that out of sitting a past paper which is just for exam practice really

But it’s the same for everyone. She’s not disadvantaged .Predictions and mocks are not final exams. All anyone can do is give their best guess as to what students will achieve. It can work both ways. My son got a C in his chemistry mock, was predicted a B in the final exam, got an A.

It's not that it wasn't good enough for the mark scheme, it's that she didn't perform as well as the others in her year group on that paper.

People get surprise low grades all the time. Some people don't perform as well under exam conditions, some people panic, some teachers set unrealistic expectations of what students can achieve, sometimes that paper just didn't suit that students knowledge, students can misinterpret markschemes and think they are performing better than they are.

All a student can ever do is try their hardest.

I think I understand better now how the boundaries work. I think dd's teachers and grinds were not on the ball and gave a false sense of security. Nobody except me seems to think that though. I mean surely you do mocks and past papers to get an idea of how you will do? Saying they are just for exam practice seems deliberately obtuse as you're saying well you can do this exam but don't be thinking that that's any prediction . Dds grade predictions were wildly out. Maybe better schools taught their students better or prepared them using the 2019 and prior marking schemes. It's just hard for me and DD to understand how you can never get lower than A or A* for 2 years and come out with a Cs. And she didn't have nerves in the exam cos we got back the papers and her grind teacher said her answers were good. Just not good enough though. It seems to be a very opaque system. But everyone else seems to think it's great so I take on board that I either don't understand or am so thick I can't get my head round it

So the thing is if you say to a kid who has never gotten lower than A or A * for
2 years that they "got what they deserved" how does that make them feel? That they "got what they deserved accost g to the rest of their peers". That's what dd is struggling with. That's why she got copies of
The papers and showed them to her teachers and her tutor and they all said if they had marked it she would have gotten an A . I know you're all saying that they don't have the mark scheme. But the point I'm trying to make is that DD produced answers that her teachers and grind would
Have given an A to. Does that make sense to anyone at all.