Active discussions

Please or to access all these features

Talk
Secondary education

Connect with other parents whose children are starting secondary school on this forum.

Original poster

Predicted Grades for UCAS

107 replies

LibraryBook · 11/09/2013 14:19

Just trying to understand how different teachers/schools arrive at predicted UCAS grades for A2, to put on the UCAS form?

I'd be grateful for any explanations.

OP posts:
secretscwirrels · 12/09/2013 17:22

What are Sutton Trust unis? I thought Sutton trust was for the under privileged?
I think the top unis require STEP for Maths degrees which is considered a better measure of mathematical ability so they will sometimes let you in if you do well at STEP but miss the A* in FM.

JGBMum · 12/09/2013 17:20

I don't think it's as straightforward as that though. If you take the example of getting an AAA offer, achieving AAB and being accepted, you risk struggling to keep up.

Bristol uni engineering require AAA, reduced to AAA if further maths is included. However, their average UCAS point score per student is 515 points (Guardian League table). That's the equivalent of A A* A at A2 and 2xA at AS level.

A student with AAB at A2, and 1xA at AS has 400 UCAS points.

Original poster
LibraryBook · 12/09/2013 16:33

secretscwirrels - universtities at the top end have a range of standard offers. A*AA or AAA including further maths, that sort of thing. If you're predicted ABB, you aren't going to get an offer. But interestingly if you are given an AAA including further maths offer, but narrowly miss, and achieve, let's say, ABB, they might let you in. I'm certain it's why independent schools have better success at admissions to Sutton Trust universities.

How dare ordinary kids have aspirations eh?

OP posts:
secretscwirrels · 12/09/2013 15:54

LibraryBook Are you suggesting that not having a part time job improves grades?
I really cannot see that it's in the student's interest to inflate their grade prediction. They might just get a higher offer than they would have done with a truer prediction.

noblegiraffe · 12/09/2013 13:18

They should go and talk to their teacher then. It's quite common for students to come and ask for their predicted grade to be put up because they need it for a certain university. Usually we ask for some sign of renewed dedication, plans to get a tutor, give up job etc, then bump the grade on the understanding that they need to make sure that their insurance offer covers their original prediction,

Original poster
LibraryBook · 12/09/2013 13:12

A smaller percentage of university grades were overpredicted 41.7% than were incorrect 48.7% so I don't get your drift.

If a teacher is wrong about the predicted grade, it doesn't matter if they were wrong due to pessimism or optimism. But it may matter to the student.

Anyway, UCAS is a bit of a dog's dinner. Luckily my son got his A grades at AS but then he doesn't have a part-time job like some of his friends who narrowly missed out on the top grade for AS and are therefore denied access to the best universities due tot their predictions.

OP posts:
tiggytape · 12/09/2013 12:13

This reply has been deleted

Message withdrawn at poster's request.

BeckAndCall · 12/09/2013 12:07

It's not an independent school thing library - I was talking about an independent school when I quoted the policy above of predicting at AS grade as the rule.

noblegiraffe · 12/09/2013 12:03

Actually, it's not 2 in 5 students, is it? It's 2 in 5 exam results. Given that students sit 3+ exams each, that means that more students are affected.

So no, I can't support your suggestion of more teachers overestimating predicted grades.

Original poster
LibraryBook · 12/09/2013 11:46

51.7% of all predictions were accurate, 41.7% of all predictions were over-predicted by at least one grade, and only 6.6% of all predicted grades were under-predicted.
? Just under 90% of grades were accurately predicted to within one grade.
? A grades were predicted most accurately with 63.8% of A grades having been
accurately predicted.
? C grades were the least accurately predicted with only 39.4% accurately predicted.

OP posts:
Original poster
LibraryBook · 12/09/2013 11:43

noblegiraffe - don't be dramatic, that's why we have a firm conditional and insurance conditional place.

OP posts:
noblegiraffe · 12/09/2013 11:35

www.gov.uk/government/uploads/system/uploads/attachment_data/file/32412/11-1043-investigating-accuracy-predicted-a-level-grades.pdf

This says that 41.7% of grades were overpredicted. Meaning that believing in a student's capacity to improve or dangling a university place in front of them isn't working, rather is actually causing stress and worry to up to 2 in 5 students who don't get their predicted grades and are possibly thrown into clearing.

Original poster
LibraryBook · 12/09/2013 11:27

hatsybatsy - but the student might get AAB or even ABB and the AAA offer could easily let them in anyway. This happens A LOT. My son's old independent school predicts a grade up from that achieved at AS (except A* which is discretionary). No wonder independent schools get their kids' bums on seats at the better universities.

Plus, it seems unreasonable not to believe in the capacity for your students to improve and to understand the power of incentive (of having a good university place offer, dangling in front of you).

OP posts:
hatsybatsy · 12/09/2013 10:41

surely that would waste everyone's time? If a kid is predicted AAA, then he will get university offers accordingly. If the prediction was 'optimistic' and he only gets BBB then he has no university place.

The predictions are meant to be realistic - not sure where your 50% accurate stat comes from?

Original poster
LibraryBook · 12/09/2013 10:20

But the universities can already see the applicants AS grade, I think the predicted grade should be a little optimistic. So as to encourage the student to fulfill potential rather than plumping for safety. That's also why students should make an aspirationsl choice of university (or more than one) on the UCAS form.

OP posts:
Original poster
LibraryBook · 12/09/2013 10:17

Predicted grades are 50% accurate, roughly.

OP posts:
noblegiraffe · 12/09/2013 10:03

In maths we would certainly not predict higher than AS due to dropping a subject, because the A-level modules are harder than the AS ones so it often evens out. Students sometimes go up or down a grade so we look carefully at how they are doing in Y13, planned resits etc.

BeckAndCall · 12/09/2013 09:59

At my DDs school, it is entries on AS results, we are told. We were told in a way that made me think there are no exceptions BUT I have no idea if they ever do make exceptions.

As much as anything, the school's reputation is on the line and they wouldn't want to be seen as deliberately overestimating. And they're pretty close to right most of the time, so far as I know.

Original poster
LibraryBook · 12/09/2013 08:52

Pretty daisies that sounds fair enough about the A* grade.

OP posts:
tiggytape · 12/09/2013 08:43

This reply has been deleted

Message withdrawn at poster's request.

Original poster
LibraryBook · 12/09/2013 08:28

Yes, of course knowing the student might help. Grin

But some schools use basic Alis which is just an average GCSE score.

DS is very happy with his predictions. But a couple of his friends are dropping from 4 to 3 subjects and the teachers are predicting the same grade they acheived in AS. Seems bonkers to me.

OP posts:
TheFallenMadonna · 11/09/2013 23:27

Professional judgement. Or a multifactorial analysis if you prefer...

prettydaisies · 11/09/2013 23:19

All DD has said is that, in her school, you will be predicted an A* if you get over 90% in your AS exams.

BeenFluffy · 11/09/2013 23:13

This reply has been deleted

Message withdrawn at poster's request.

Kez100 · 11/09/2013 21:49

Professional judgement is a darn sight less vauge than some statistical model which creates an average!

There are only 7 A level grades. Knowing the student and having some AS scores to hand - I'd say professional judgement to pick the grade from those 7 would be most accurate and not in any way vague. It's judgement not sticking a pin in a list.