Contextualized GCSE score

[Perzival]

Hi, I've been looking for info on how Oxford calculate the cGCSE score but can't find anything definite. I was just wondering if anyone knew please?

I'm going to see if I can find any info from 2019 and 2020 to look at cgcse as these weren't included in the above foi.

@Reallyunlikely if you look at the third data set it lists how many of each grade the students at each school got. You can search for the school you wNt to look at.

Do those tables tell you the actual number of 9's, 8's etc, I can't see that. Or do you just take the attainment 8 scores and extrapolate from that?

The spreadsheets do.

If you go to https://www.compare-school-performance.service.gov.uk/download-data?currentstep=datatypes&datatypes=gias&datatypes=ks4&datatypes=ks4underlying&downloadYear=2021-2022&la=0&regiontype=all and download the CSV or XLS file, you'll be able to open it in Excel (you'll need something like 7-Zip to extract the files from the archive).

If I filter on my old school (underperforming comprehensive) and maths, say, I get this:

I imagine, though can't be sure, that Oxford will calculate the cGCSE standard deviation using the distribution of GCSE grades from the applicant's school (maybe will some cut off for number of GCSEs taken) centred on the expected number of A*/9/8s using the mean or mode.

So unless you know the cut off assumption I'm not sure you can calculate this yourself. But I would have thought that for most non-selective state school applicants they will have an average GCSE score of 8.5+ against a school average of around 5 (the national average for state schools)

@tadger98 thank you for the spreadsheets. I looked at my ds's school and struggled to work it out if I'm honest.

I think most students at his school took 10 gcses. Roughly 50% of the grades for each subject were an eight or a nine. If ds got six 8/9's out of ten subjects so 60%. Would it follow that he'd more than likely have a positive cgcse or at least a very low negative? I appreciate this isn't a straight ask, am I right though that it's unlikely he'll be -2 or -3.

Ahh, 3rd dataset, that was where I was going wrong. Thank you.
Although, as tadger says, I guess the simple answer is that at a very average comprehensive, and ours is pretty bang-on average by most metrics, you will be comparing against an average of 5's.

@Perzival Yes my understanding is that your DC would get a positive cGCSE. But given the features of the distribution of grades you provide I doubt it would be much different from zero. I think the way to think about it is that the cGCSE allows Oxford to select non-selective state school pupils more than would otherwise be the case from their GCSE results and the admissions test. I'm not sure it's meant to distinguish between students at selective state and independent schools with high average grades like those at your DC's school. That's not what Oxford mean by contextualised as far as my understanding goes.

It’s not worth overthinking it. You will not be able to work out the cGCSE score as such- the method used is a secret and if you look at FOI requests you will see that Oxford refuse to reveal the algorithm that is used (excuse given to do with benefitting their competitors). It is certainly not as simple as you getting ‘+1’ if you got one more 8/9 than the average for your school.
The most information I found some months ago was in an admissions manual for one subject, which had been released as part of a FOI request. I can’t remember the subject and can’t seem to find it again. But…. it was a straight line graph. It plotted the percentage of 7-9s gained in a school, against the ‘expected’ number of 8/9 grades an Oxford applicant from that school should have. Again- can’t remember the exact details, but if I remember rightly, I remember my son’s school (a good comp) got 38% grades at 7-9 in his year of GCSE. The graph showed that an Oxford applicant with a cGCSE score of 0.0 from that school would be expected to have achieved something like five grade 8/9 grades (might be remembering that number wrong!) So those that had more than five 8/9grades would have a positive cGCSE score, but the exact equation for how that score would be calculated had been removed. Obviously , those with fewer than five 8/9 grades from that school would have negative cGCSE.
As you would expect, the graph had a positive gradient, so by the time you got to a school that had 90% of grades at 7-9 then the ‘expected’ number of grade 8/9 for an Oxford applicant might be more like eight (example figure- really can’t remember exactly, and can’t find it again!)
Sorry for long post- feel free to scour the internet to find that graph. But not worth it.
Yes- the xGCSE Score is part of it. But every year some people with perfect 9s at GCSE don’t get in and some with a more mixed set do- it’s not all they are looking for.

I definitely agree with what @Literallynoplan says. At the end of the day Oxford will look at all the information they have on each candidate to make what must be incredibly difficult choices between lots of candidates who they know would do really well but can't accept because they haven't got enough space. But I think it is important to understand the different weights across the assessment tests, GCSEs and interviews. At least for physics that tells you that preparing for each of the 2 or 3 interviews is really really important.

Thank you all for your input. I suppose the cgcse score is the one part of the application that he can't change now whatever he does. We're very proud of him regardless of what happens.

It's been eye opening reading the school results and also looking into the application. We don't know anyone that's been through the application before so we're trying our best to get a good understanding of how it works. Hopefully he won't get a negative cgcse and will do well on the pat and the interviews.

I'm really grateful for all the help. Thank you.

I saw this thread and wanted to correct something about cGCSEs which may not be clear. cGCSE compare an applicant's GCSE grades against those of other Oxford applicants from "similar" schools. They do not compare against the average GCSEs inside an applicant's school.

So cGCSE for an applicant from a 'standard' comprehensive comes from comparing their GCSEs to that of other Oxford applicants from 'standard' comprehensives.

Most applicants have between -1 and +1 cGCSE. Something much above +2 probably needs 11 - 13 8/9s* *from a very ordinary state school. Something much less than -2 probably needs 1 -2 8/9s from a strong school which sends lots of kids to Oxbridge every year.

Thank you @anonacademic123 ds' schools sends 1-3 a year but is rated inadequate by ofsted. Will that make a difference please?

If it is inadequate inadequate, if you get my drift, then this will be reflected in the comparator schools used for cGCSE and often will also be mentioned explicitly in the school reference. Tutors then take all this information into account when evaluating an applicant.

If it is inadequate because e.g. it failed its Ofsted on safeguarding but is otherwise academically good with generally decent results, then things might be different.

It is not uncommon for tutors to look up an individual school and its website etc to get a more detailed sense of the school than just a name and type.

I really wouldn't over think it. Your DS's (excellent) GCSE results can't change upwards of downwards. He can influence his predicted grades through hard work, along with his PAT. Don't worry about things you can't change.

Thank you both. I think the school is good, it was rated good for education and inadequate for the others. It has previously been outstanding. Educationally it does well.

@Dearover i posted this a while ago and came to that conclusion. I just want to understand the whole process- it isn't something me or dh have ever done and ds is sn amazing kid. I just want to make sure we help him as much as we can.

Does anyone know how is works for people who apply with their A level results in hand? So, if your GCSEs were good but maybe not good enough according the algorithm, but you get A*x3 at A level. Do they still care about GCSEs?

Just make sure his supra curriculars are good and he practices for the PAT.

I doubt Oxford would consider any more than 9 or 10 GCSEs as the vast majority of schools (certainly non-selective state schools) are not resourced to support a student taking more than that.

Of course the other obvious point is that foreign students will get a score of zero.

@Lassi all information matters, but in subjects with tests (e.g. physics/maths etc) then the test score is the most importance piece of information prior to interview.

Tutors are human and most have kids themselves. It is true that most successful Oxford applicants have strong GCSEs across the board. But tutors understand that personalities differ and some kids are more single-minded; every year there will be kids getting in for physics with e.g. Bs and Cs at GCSE in English, languages etc. Also applicants who are e.g. dyslexic might be brilliant at maths/physics but not great at essay subjects and with very spiky GCSEs.

Step 1 = apply. Step 2 = do his best in the admissions tests. Step 3 = hope he gets an interview