Exam data nerdery

[noblegiraffe]

I'm fairly sure that there are other nerds on this site so I thought I'd post some stuff I'd collected in case anyone else is interested. Also, if you see something in the data, maybe we could chat about it.

First offering is this widget from Ofqual. You can look at the grade distribution for various subjects, or you can see how grades in one subject affects grades in another subject (the link between maths grade and computer science grade is quite pronounced, for example).

https://analytics.ofqual.gov.uk/apps/GCSE/9to1/

What I noticed is that the grade distribution for English language is weird. And there are far fewer top grades awarded than for Maths. Why? (This is with the resit data stripped out). It's the same cohort that sits both so it's not about prior attainment.

English Lang is completely untiered and everyone does it!

So does that mean that every year 2.9% (roughly) of eng lan candidates get 9s and 13% of chemistry candidates get 9s and the boundaries for marks move around to enable this? So many parents on here seem to be saying that their children have been predicted a 9 in lang but only got a 5/6 but is it just that other kids around the country maybe did better than expected and pushed the predicted 9s downwards?

That does seem to be what various data suggests. It conversely also pushes bottom upwards, so producing more 3s.

This is fascinating! Have I understood correctly that more kids did better at Eng Lit than Eng Lang? I’m assuming kids usually take both? When I did O levels only the brightest did Literature.

I’m blown away by the fact that kids do better at Maths than at English - maths has such a reputation for being disliked.

Are the % grades for A level similar across all boards? That is, is the AQA chart posted pretty representative?

It's always generally been the case that more kids do better at lit than language.

I think it should be shown much earlier maybe year 9 when they have started their option subjects. Also the expected parental contribution for uni too with salary bands and smelling salts to revive parents after they faint I think parents should be informed early on.

I had a Cambridge Assessment report from 2010, not sure if there is a more up to date one that showed the progression from GCSE to A levels grades wise. Here for any other nerds https://www.cambridgeassessment.org.uk/Images/153531-progression-from-gcse-to-as-and-a-level-2010-.pdf

@JustHereWithMyPopcorn As published above maths is a hard A level from a 7 at GCSE and you can see why. I am sure @noblegiraffe will correct me if I am wrong but DC's sat OCR maths, we were told that 50% of the higher paper determines grades 7-9 meaning you can achieve a 7 without potentially having 50% of the maths knowledge secure This then makes it harder at A level. Both of my sons had solid grade 9s and have found A level maths relatively easy because they didn't struggle in class. They also did/are doing further maths A level too.

My DC's school pick and choose their exam boards, OCR for maths and computer science, Edexcel for Statistics, Educas for English Language and AQA for everything else.

@MrsHamlet do you feel there should be a foundation paper for English Language? Mine did really well but I can imagine it is soul destroying for some children to have a paper that covers all ranges.

If there were to be a foundation paper, we'd almost certainly end up back to setting - which I am not for. There's no easy solution.

I think there must also be something in the fact hat in maths you can choose not to teach all the content, this freeing up time for reinforcement of weaker topics . In every non tiered subject you have to teach all content - there's no special extra for grades 8/9 and nothing you can't teach.

In theory only in my head this might man more very top grades in maths and perhaps a better chance of getting grade 4 because you are just taught no more than you need to know.

Conversely, I can see it might hold back some students who, because of the set they are in, perhaps, don't access content they could actually grasp.

I think some of the kids in my DD's separate sciences group actually did foundation rather than the higher tier. From what some of them said to DD they thought there was a chance that they'd at least get one science even if they failed the other two whereas combined might mean they got none because the really bad ones would drag the other one down! Maybe that muddies the water science wise because obviously everyone has to do science in some form and the stats don't differentiate between who took foundation and who took higher.

In theory only in my head this might man more very top grades in maths and perhaps a better chance of getting grade 4 because you are just taught no more than you need to know.

But the chance of getting a grade 4 or 7 is pre-set. I don't know at what point it was decided that kids were better at maths than English.

But thinking about it, kids are better at maths than English, because boys massively underperform in English so the grades should be lower.

No consolation for English departments who will be compared unfairly to maths departments if both perform perfectly in line with national expectations.

Yeah, I just saw that data about bell curves and how thry squashed top grades down more thus year. Need to look it out.

If memory serves for the separate sciences about 1 in 8 entries are for foundation tier. Whereas for what used to be called dual award around half of the nearly 3 times as many entires as there are for separate sciences are foundation tier.
So I doubt what you describe is a very large effect.

Interesting as my eldest two were the other way around.

And why do boys do worse at English? Any theories? I’ve 3 daughters but just curious.

@MrsHamlet I’m really interested to understand your rationale for not setting in English please. As a kid we were set but at my daughters’ school (I’ve 3) they change from setting to not and back again. The sets do different texts too. Not sure if middle daughter’s unset year all had same texts or not.

Yes, since it became compulsory to do both since the Gove reforms. It used to be the case that many schools did not even enter less able students for Lit.

IMO students do better as there is a lot of content they can learn, and provided they know the texts very well (with the except of unseen section) then they can sort of learn the mechanics of the essay and do well. Whereas for language, ALL of the reading reading texts are unseen and often very challenging. The students have to respond to them in the moment and the questions make them jump through stupid hoops. I think the Lang spec is due to be reformed and should be better prep for a level, whereas lit I feel is pretty good and actually an improvement on previous specs where you could get a GCSE in Literature without having to read a full novel!

English is not set at my children's comprehensive, it never has been.

Also I think they should accept a pass in Eng Lang or Lit to not have do re-take.

I am pretty sure that this is already the case. The trouble is lots of post 16 providers, employers , and universities, stipulate a 4 or above in English Language GCSE.

Who decides that the top 31% in Physics A level get A or A star but only the top 22% in computing are awarded these top grades ?
And is the reasoning that it is better calibre students taking Physics than Computing ? If so, how is this worked out ? From GCSE grades ?

https://www.gov.uk/guidance/16-to-19-funding-maths-and-english-condition-of-funding

If you want a more extreme example, see the 2% A* in A level media...

The grade proportions get 'calibrated' every so often but, to my knowledge, this has not been done for media or film yet, post A level reform, I assume partly because CAGs got in the way. My understanding is that this is acknowledged to be an issue but that the pledge to bring things 'back to 2019' (which doesn't really work very satisfactorily for the last tranche of reformed subjects) got in the way of any adjustment of grade distribution. And yet the exam board has sad they know students taking these subjects are adapting well to the reforms , the subjects are growing, the teaching is improving, and that there is a general acceptance that the students are more able.

Philosophy teacher here! Simple answer is... they don't. Half of the marks are for those essays, and they are always low even for good candidates. A* was 76% this year, and most of that will be picked up from the short answer questions. Admittedly, those are also awful; they ask for definitions which are not given in the spec and there is no board-endorsed textbook. Honestly, the marking has been so inconsistent since the spec was reformed in 2019 that I'm thinking of switching to RE (which really goes against all my principles), having taught Phil for over 10 years.

Published philosophers would all fail the A-level, you need to write in a very particular way to hit to mark scheme criteria. But that's true of a lot of subjects.

I wouldn't say that the highest level descriptors in RS are any easier to hit either. The students need to be able to write in depth about philosophy, ethics and a religion. Again, if you look at the wording of the descriptors they are ridiculous. I honestly wonder what the difference is between what is expected of a GCSE student and an undergraduate. The reforms were clearly pushed through by those intent of setting a demanding academic agenda - fair enough - but seem to have forgotten that they were writing GCSE specifications, not university degree courses.

It is true that it is possible for students to pick up enough marks on the shorter questions, but it is a shame that so many really, really able students struggle in the essays. Honestly, what is the point of having the top one or two levels if they are all but inaccessible. The jump from the old specs to the new one was massive. All credit to RS teachers who worked their socks of to deliver the new specs, in many of the options with little or no support material from the exam boards.

If you look at the new text books and compare them to the resources for the old specs it is clear that much of the new content is simply inaccessible to many students. There is so much content now that the new books are crammed with text, dozens and dozens of scholars and incredibly complex specialist language. Then bear in mind that as RS is compulsory many schools are putting their whole cohort in for the exam.

My kids grammar dies ocr.They say their pupils do better on harder papers with lower grade boundaries and think ocr is that.

There was an online backlash against the A level film reforms which did actually see the exam board change some of its plans. A professor of film called the original plans 'an undergraduate degree in two years'. Fortunately, this led to streamlining - at least of content. But students still need to study something like 15 texts in detail, which is way in excess of eg English Literature.

By way of illustration of part of why this is.
On a 10 mark question, if as a marker I am out of tolerance if the mark is 2 or different from the seed if I give the answer 10 marks, the only other mark that I am within tolerance for is 9. If I give the answer 9 marks, then if it is 8,9 or 10 I am within tolerance.
Does this mean that unless I am really certain it is worth 10 marks I err on the side of giving it 9 to maximise my chances of being within tolerance.

Piggy - that sounds very like what happened in RS. If you look at the GCSE and A Level specs now the same topics are on both. GCSE has become 'slightly lighter' A Level. Crazy.

The old RS specs were great - full of content that the students could really engage with.