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.

I love these sorts of threads.

One thing i dont quite understand is further maths gcse. My son did it, but it seems only people who look to get 7-9 in maths sit it. So if they stick to the top 5% get a 9 patterns in other subjects does this mean the bottom 30% of further maths get below 4/5 even though they are the top 30% in normal maths?

Also same with A level - if only the top grades can do it compared to other A levels dont you end up with a grade distribution all based on really good at maths people so its a harder A level?

I love that aqa one. I'm personally invested as my DC got a very high grade 9 for literature but only just scraped a 7 for language which we were surprised by. You can set it just for a grade 9 in one subject then see how many of those got each grade in a different subject.

Not sure about the comment, kids need a 5 more than a 4?? Mumsnet offspring notwithstanding, many, many students will not be doing A Levels after GCSEs and it's more a 4 needed surely? 5 GCSEs at 4-9 including maths and English is what's needed for so many post GCSE routes. Some colleges even take less as long as you have the 4 in English and maths.

Great link though @noblegiraffe . Will be perusing with great interest!

I also have to remind parents that there is also a difference for subjects that the whole cohort do, and those subjects that students choose to do. And other factors which might impact on the number of highest grades achieved. It is hardly surprising that subjects such as Polish or Russian get amazingly high grades in a school with one or two native speaker entries per year.

If it's only bright kids sitting the exams, then a higher proportion of them get the top grades - if you have a look at Classical Greek GCSE, 54% of kids who take it get a grade 9, because the kids who take Classical Greek are likely to be pretty clever.

It's the same at A-level - you can see that 41% of kids who take maths get an A/A - that isn't because it's an easier A-level, but because the entry requirements are generally around a 7 at GCSE. The A-level with the highest proportion of A/As is Further Maths.

Put in the *s next to the As as appropriate. 🤦‍♀️

Thank you @noblegiraffe that makes sense. In my head the exam was getting harder and harder as only high achievers sat it and they had to fail someone.

Exactly Noble. And subjects like mine who get pushed by the HT to take any Tom Dick or Harry who scrapes just enough GCSEs to make it into sixth form unsurprisingly have a higher number of students achieving at the lower end. The HT just sees the £££ they will bring in, not the struggle those students will have trying to manage the A Level courses.

Again, something that governers and parents can't always seem to grasp.

You don't need the highest marks to get a grade 9 though, you just need to be in the highest percentage of candidates.

To get a grade 9 for AQA English language you needed 76%. I don't know how that fits with these level descriptors?

No you don't need to reach the highest level in all of the questions on the paper to get the top grade that is very true. I just wonder when I read them if the people who write the level descriptors have ever met a 16 year old, let alone taught one.

We were battered on language again this year. At the pulling our hair out stage cos we genuinely don't know how to fix it.

And like pp, lit were 2 or even 4 grades higher. Interesting that's it's also AQA.We're with anther board.

More data in this BBC article, you can see what percentage of grades in your area were 4+ https://www.bbc.co.uk/news/education-66476095

Takeaway is that Rutland (wherever that is) is absolutely stuffed with bright kids.

Also really surprised with English language the most common grade being a 3 with 153,000 children getting a 3. That's a lot of re takes. :-(

Exam statistics for Further Maths GCSE - 20% of kids who took it got a 9.

https://filestore.aqa.org.uk/over/stat_pdf/AQA-L2-CERT-STATS-JUN-2023.PDF

Rutland has three state secondaries and two Public schools. All those kids with Prep timetabled throughout the week has to make a difference.

Takeaway is that Rutland (wherever that is) is absolutely stuffed with bright kids.

Likely to be because it's a small sample size (4 secondaries I think?) One of which is Uppingham, a selective, so that skews the data!

Tiny county with 2 selective private boarding schools!

That's the data including resits. If you filter by kids aged 16, the most common grade for English language is a 5.

It does show demonstrate how many resit kids get a 3 though.

Here you go! https://www.eduqas.co.uk/media/uzajal25/eduqas-gcse-provisional-results.pdf

No idea who sits Eduqas exams but their maths results are awful, only 27.9% of kids pass??

I’m not a teacher nor a data analyst and I didn’t think I was a nerd but maybe I am because I find this all so fascinating! Also with a DS about to start yr 11 I’m getting very nervous!

DS is in a grammar school and predicted a 9 for maths but is in set 2, set 1 sit further maths gcse. The school only let pupils do maths A level above a 7 (really they only want 8-9). Half of the A level students (approx 100) take maths A level with high numbers usually achieving A*-A. They certainly seem to understand that very clear line of high grade gcse = high grade a level in maths which doesn’t necessarily happen in other subjects. Maths is the ONLY subject they set for.

No idea who sits Eduqas exams but their maths results are awful, only 27.9% of kids pass??

Eduqas maths is 2x 135min exams (1 calc, 1 not) vs other boards 3 x 90 (1 non-calc , 2 calc). Not sure that's enough to make such a difference though.

Welsh board usually have the lowest grade boundaries, so I think schools choose it specifically when their cohort is lower ability/attainment or is significantly disadvantaged.

I've been doing Welsh board for years. It's not done us any favours.

Did AQA before that and thought that was harsh. Kids used to really struggle with the structure question, but I think the descriptive writing task o AQA was easier than the story task on Eduqas.

I do wish we could just have one exam board; mainly to stop the 'what if we change board?' Conversations every year.

Thanks for this noble, I tutor for science and I've noticed my 'top' students have achieved their 9s but my 'middle' haven't have done as well as I would have predicted, so I'll be having a look when I get a moment.

I do wish we could just have one exam board

Gove tried! i think there was some issue like Edexcel would have to get the maths gig and AQA would have to get the English gig and the government handing government contracts directly to private companies is bad. Or at least it was considered bad back then.