Predicted Grades for UCAS

[LibraryBook]

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.

[friday16]

This is a good argument for Y13 sitting exams in April, getting results in June then applying to uni.

Or later; another proposal was to delay Freshers by a couple of weeks, take advantage of more IT (as you say, the process is now all online so should be quicker) and do it from exams in May, results in July. I did read the universities' response the last time it was proposed, but I can't now recall all their objections. Running a full admission process over the summer would be difficult for universities that interview, of course.

[Cerisier]

This is a good argument for Y13 sitting exams in April, getting results in June then applying to uni. In the days of internet and computers why does the same system still hold that we had with UCCA years ago?

Without AS exams there would be 5 terms of teaching, which is probably more than they have now.

[IceBeing]

My personal experience is that indy seem over optimistic and state are pessimistic...

I saw people with UMS grades above 95% that hadn't been predicted A star (90%)...again only from state...I don't know why that happens...

Institutes of physics (IOP) know perfectly well that equally achieving women get lower average predictions than their male counterparts....but again there is no way we can do anything in admissions....we certainly can't start positively discriminating.

[titchy]

Interesting, although you presumably that observation is only based on the very top applicants given your standard offer. Do you know if it's true of lower grades?

Would you say state predictions are more accurate than indi ones?

[IceBeing]

bollocks the bold system screwed up my message!

What I mean is that people predicted triple A star from an indy school actually get the same number of points as people predicted A star, A, A from a state school...

[IceBeing]

Yes we can make contextual offers..to individuals who seem to have potential beyond what they are currently achieving. But that is a one off thing and something to be considered when the school is really bad or there are other mitigating circs.

The real problem is looking at all the people being predicted AAA versus those being predicted AAA etc. We can't assume that a triple A prediction from and indy school is worth the same as an AAA from a state school even though they may well be in reality.

It is a known and accepted fact that teachers under predict more for women than men in physics. So not really interested in any reasons why that wouldn't be true. It IS true as found by comparing predicted grades with achieved grades.

Similarly I saw several predictions from indy schools that were algebraically impossible given the AS grades....not any from state schools. So not everyone is predicting directly from AS grades....

[DalmationDots]

My DD was at a top 10 girls academic independent school, her AS grades were AABB (later -post predictions- one of the Bs was remarked and changed to an A)
She was predicted AAC, got 5 russell group offers for courses which had entry requirements of AAB. So the schools (at the time realistic) prediction of a C did not seem to harm her. She came out with AAB in the end.
When on the course she has found her and her friend's offers varied greatly. Some being offered BBB, others offered AAA or AAA.

DD asked her tutor at uni out of curiosity and was told the tutors adjust considering what school you go to (e.g. if it is a poor state school in a deprived area your offer will be lower), how much they want you, other circumstances (DD had personal family circumstances which impacted her AS exams causing the predicted C grade). Her uni and course is great on accessibility and recognising the importance of a varied set of students. Other unis may do things a lot more systematically and 'to the book' e.g. all AAA offers.

[BeckAndCall]

I find your analysis fundamentally flawed, icebeing for a number of reasons including:

As an admission tutor at an RG university you do have the ability to make contextual offers if you feel grades are over or under predicted in one group of applicants - in many courses the stated admissions criteria shows a range;

Your argument is based on the assumption that independent schools DO over predict - and several people on here, myself included, have illustrated that their schools use only the AS grades as a predictor and do not inflate - so there is no cross sector pattern;

Your argument that a male makes stand more chance of getting on a physics course flies completely in the face of the Women in Science initiative which is moving the dial on that statistic - which is in any way majorly affected by the number of applicants in the group, rather than differential offers within the group;

And it ignores the fact that many students in independent schools are already predicted to over achieve the offers made to them ( eg achieving A*AA against an AAB offer) whereas the average achievement in state schools is lower, so students arriving from state school are more likely to have hit the offer spot on. So that's why they have lower average grades - not because it was harder to get an offer in the first place.

But is is very hard to generalise about 24 universities offering thousands of courses to tens of thousands of students, so anybody else's take on the situation is just as likely to be valid as mine. And there's a whole industry out there trying to analyse all of these things one way or another and I just don't accept that it's clear cut.

[IceBeing]

I said that students admitted to RG unis have different grades depending on what sort of school they came from. The implication being that it is 'easier'to get in if you come from an independent school. It is also 'easier' to get in if you are male and white (to physics anyway).

This has nothing to do with the difficulty of the offer (we offer AAA to everyone essentially) it has to do with us being more likely to select people with higher predicted grades. Given a bunch of students who all actually got AAA, the Indy students had higher predicted grades than the state school...the men had higher predicted grades than the women and the ethnic minorities had lower predicted grades than everyone else.

This is likely at least in part down to bias on the part of teachers predicting grades (either conscious or unconscious).

Does that make it clearer?

[noblegiraffe]

Why are you asking? I wondered why you asked me if I teach further maths earlier!

[LibraryBook]

titchy - I'm just wondering what it is you teach?

[titchy]

If the rest of your post is directed at me I agree with you RG entry standards (and tariff scores of entrants) is much higher!

Not sure why my (or any other posters') degree or alma mater is relevant.

[titchy]

Icebeing said RG students have lower tariff in her15.52 post.

[LibraryBook]

titchy - what ARE you talking about? NObody said RG have a lower tariff. Of course the better unis have higher entry standards. But it is possible that someone could have 750 UCAS points (more than most of those doing medicine at Oxford) and not have a hope in hell of receiving a RG offer let alone an offer from Oxford. You must get that? What is your degree from by the way, and what subject?

[LibraryBook]

Dominodonkey - sorry.

[Dominodonkey]

Please don't make unfounded accusations op. I have only ever had one user name and this is it.

[titchy]

Actually look at the league tables and order by entry standards, Oxbridge imperial etc right at the top!

[noblegiraffe]

Indy schools are 'better' at predicting grades not because of any special skill but because they predict more A/A* grades due to selective intake and these are easier to predict. State schools are good at predicting the top grades too.

[titchy]

Library the THES and Which may want to discuss lowering SNC to BBB but HEFCE have been told not to, and until we get a change in Gov and policy BBB unlikely.

RG having lower tariff than others - haven't seen the data (source?) but RG applicants don't have as big a need to put down all their grade 8s. They may well also take more students with non-tariff qualifications - overseas or the pre-u for example.

[LibraryBook]

Independent schools are actually more accurate at predicting grades than maintained schools.

And UCAS tarriff is not a very good guide. An A in A level further maths for example is worth the same UCAS points as an A in general studies when you are implying to Imperial for Physics.

RG universities don't admit using UCAS points tariff, normally, they ask for specific grades, often in specific subjects.

[IceBeing]

Not really sure what this thread was aimed at but I do have a question for teachers....

Recent stats show that students in R group unis have a lower average UCAS tarrif than students from state schools. The obvious implication is that indy schools over predict more than state.

Similarly there is good evidence that gender bias, and race bias occur (eg girls doing physics don't get the same predictions as boys and ethnic minorities do badly for predictions across the board.

The question is what can be done about it?

We (admissions tutors) can't ignore predicted grades, we can't bias against students from indy schools our hands are tied....so how can we fix this and start getting better diversity in RG unis?

[LibraryBook]

friday16 - The entire Russell Group didn't go into clearing. The following Russell Group universities had no need to enter clearing in 2013 Ox, Camb, Imperial, LSE, UCL, Bristol.

[LibraryBook]

friday - eh? That is what I've been saying.

[LibraryBook]

titchy - yes, I meant ABB, thanks. But there was some discussion about it being lowered to BBB.

this is worthwhile reading

and this

[friday16]

"I'm just saying it would be absolutely absurd if grades fell hugely and universities were thus unable to fill their programmes"

Why do you believe that a fall in grades would result in universities being unable to fill their programmes? Do you believe that universities are unable to, and indeed do not (a) offer or (b) confirm at below the general offers listed in the prospectus?

Suppose, for the sake of argument, that A2 results dropped by two grades between now at October 2014. Do you think that (a) universities would just confirm places for students holding firm offers but who missed their grades (b) universities would take unfilled places into clearing and accept whatever grades were going or (c) universities would shrug their shoulders and lose huge piles of money? Reality is (a) for universities holding a lot of firms, and (b) for universities holding a lot of insurances. (c) will never happen again. The situation with Southampton (and others) last year was because of misplaced machismo about "we're so popular we never go into clearing". This year, the entire Russell Group went into clearing, and announced it before results were published.