Predicted grades - should we move away from aspirational, and move towards realistic

[Aragog]

I get the impression every year that the aspirational UCAS (and to a lesser extend gcse) grades cause issues for pupils, and lead to far more disappointment than the actual grades awarded do.

This year has really highlighted that.

To a 16-18y a predicted grade is just that - they believe them to be the grade they'll get. The fact that in most cases they are 'aspirational' totally bypasses most of these young people in my experience. A hugely,high number of young people achieve lower grades than their UCAS predictions.

I believe they can do more harm than good, and ultimately leads to a feeling of disappointment and failure in results day.

Should we be looking at giving students either:

(a) realistic predictions - ones they are 'most likely' to achieve based on their performance in year 12 (or 10 if gcse) and updated after mocks/coursework during the year if need be.

Or

(B) both realistic and aspirational grades - at A level this might help with selecting their firm choice (usually aspirational) and their insurance (usually realistic)

At the very least we need to do these young people a favour and explain much more clearly that the grades given are aspirational and they only have a chance of achieving them IF they pull out all the stops on the day.

On MN every year, as well as in real life, we see teens and parents alike commenting that the child didn't achieve their predicted grades, and how the child feels like they've 'failed' even if their actual grades are good ones,

Is it time for a change of approach?

@NotDonna

A pupil predicted AAA who gets AAB will potentially end up in a different place than one predicted ABB who then gets AAB. But it is the final grades that should count. Predicted grades are a foot in the door - the pupil predicted ABB can't apply to the same places.

Unconfident students will look at universities at the lower end rather than the top end of their range.

The WP criteria are really narrow, they are intended to look like something is being done without disrupting things too much imo.

Here's a recent press write up of some recent relevant research www.glasgowtimes.co.uk/news/viralnews/18641678.university-admissions-need-overhaul-avoid-inaccurate-predicted-grades/

A quote: High-achieving students in non-selective state schools are also more likely to be under-predicted at A-level compared to their grammar and private school peers

I have experience of both private and state schools. Private schools are very good at the admissions game!

There are perverse incentives for schools regarding predicting accurately.

As already said, once a student has an offer (achieved with an inflated prediction) even if they miss by 1 or 2 grades, even good Unis often take them anyway. So getting the place can be more dependent on the predictions which determine the offer than the grades themselves.

This year, many of next year's places have already been given away. They have been given in the 2nd phase after CAGs were used to students who probably wouldn't have achieved their CAG - the algorithm downgraded them and even if it wasn't fair on an individual basis the algorithm gave the standard yearly distribution of grades which the Unis base their offers on, so they can roughly get the right number of students for places. So next year, the better Unis will be finding a proportion of their students (from this year) are probably of. Lower standard than usual. This, plus the fears that there could be disruption to exams again and use or partial use of teacher grades again, mean Unis might require higher grades for offers. This in itself will mean teachers might need to be even more generous than usual, to ensure their students get offers. There really is zero incentive for teachers to be more realistic and it wouldn't help the students get the places......becaue there would never be a way of ensuring all teachers were realistic. It's an example of the Prisoners Dilemma - you can't rely on others doing something that might benefit the whole group, so you have to act to benefit yourself, which in the process might lead to a 2nd best and not optimum outcome for everyone.

At the same time, no doubt teachers everywhere will be telling this year's exam classes to bear in mind that all their work really counts and might be needed to determine any teacher grades....hopefully encouraging harder, long term consistent work from students, which can only be a good thing in terms of improving their finishing point of skills and knowledge.

High-achieving students in non-selective state schools are also more likely to be under-predicted at A-level compared to their grammar and private school peers

Really because I read states over predict (or at least what they predict and what their students get is a bigger gap than seen at private schools - though you can imagine a number of reasons for this).

Also numbers in non-selective state are higher than grammar or private. Please don’t let us get back to grammars are the way forward. What got working class kids into uni was comprehensive education that let them sit O and then A levels.

Also huge amounts have been done to widen access and there are lots of criteria in play that allow substantive grade adjustments for kids from poorer backgrounds, and other factors (carers, care leavers) and evening then further adjustments happen on results day.

But agree post results applications would be better.

many of next year's places have already been given away. They have been given in the 2nd phase after CAGs were used to students who probably wouldn't have achieved their CAG - the algorithm downgraded them and even if it wasn't fair on an individual basis the algorithm gave the standard yearly distribution of grades which the Unis base their offers on, so they can roughly get the right number of students for places. So next year, the better Unis will be finding a proportion of their students (from this year) are probably of. Lower standard than usual.

Unis have taken as many extra as they can - I think deferrals will be lower than expected.

We adjust fervently in the background each results day across the sector - We did this year with bells on for ofqual. A levels, after modular a-levels ended, and a bit of a re standard setting of results and reduction of coursework have been deflating. But if course top tier unis still keep asking for the 3 As - you can normally take as many 3 A students as you want so it became a great way for Russell Group to make some money (it was going to be capped this year so as not to destabilise the sector if admissions were down). So requirements have remained high and schools predict for the requirements and then the unis adjust where they can on results day and shove surplus places into clearing with lower requirements. It’s a self perpetuating cycle.

So we don’t need to worry that they will be this unusual situation coming with ‘lower standard’ students.

But agree the algorithm gave the distribution but we’ve been adjusting on that to get people in for the last 3 years.

Excuse my ignorance @Gwynfluff but why would Oxbridge hate post A level admissions? I can't understand how any university would prefer all the unknowns that come with the present system. I've long thought post A level applications is the way to go and am really hoping that Covid could provide an opportunity.

Can't first year of university just start a month later than the other years? End of October or something?

Because Oxbridge have entrance tests and interviews so they will have to readjust that schedule.

Really because I read states over predict would be interested to read this if you have a link as not my expert or what I have read, and the precise opposite of the link above.

Private schools are ruthless at admissions support.

Sorry, expert = experience

Private schools are ruthless at admissions support.

And their kids get their grades more often - but that’s not surprising. Though state school kids tend to perform better at Uni (had to be more self directed to get their grades).

Link here but it is from 2016
www.ucu.org.uk/media/8409/Predicted-grades-accuracy-and-impact-Dec-16/pdf/Predicted_grades_report_Dec2016.pdf

Perhaps there has been a change.

@ChanceEncounter

Read it again - it does seem there is a small group of disadvantaged low achievers who get under predicted. Also think there is some subtlety between over/underpredicted and ‘accurate’ prediction. It’s hard off teacher assessments and statistical modelling.

Also as performance between gcse and a level is quite a spiky line (more so in disadvantaged), if you control for prior attainment then, states under predict. So the student overshoots then target as the prior attainment is less relevant than you might imagine in disadvantaged. At least I think that’s what she means as she does say they overpredict!

I agree. This totally pissed me off. I hadn't realised before. I felt like I'd been misled, like the relationship between me and the school, the teachers, was almost a 'sham'.

I had a long discussion with deputy head on the phone about DS ones GCSE results.

He did well with 8's, 7's, couple of 6's and a 5.

The 6 was for his Eng Lang, Which he is doing A-level and was predicted and 8 possibly even a 9 if he really pushed himself. And the 5, RE predicted an 8.

Every parents evening, for 5 years, I've gone, them saying he is very bright, he could do better if he really applied himself .... he's getting solid 6's, 7's ....

This is what I took it to be! I saw his target grades, his predicted grades. Thru told me what exam result he got in every end of year test.....

He got a 4 in mocks, and I was like 'what's going on here, no parties for you! Get studying!'.

I did my parental duty and then when I'm speaking to the deputy head he poo h-pooh's all this?

and I'm left feeling like I don't know what's happened through the whole of the schooling.

What's the point?

@Oblomov20 are they letting him take the Eng Lang a-level?

In some states they are more cautious with Predictions and tightly mark the mocks to make students put the work in - which is the it he missed.

Someone was asking about other countries. I know two systems (grew up in one, live in the other).
In Belgium you finish your secondary school eductation, there are no national exams. There are four types of secondary school, leading to four types of secondary school diploma (academic, technical, arts, vocational). If you have any of the first three you go up to the university you want to apply for when admissions open (usually August to start in September the same year) and just sign up for whichever course you want. The exceptions are medicine (has an exam in July or August which you need to pass and take proof of passing along with you to sign up) and arts degrees like music which have auditions. If you've done a vocational secondary school course, you can still go to university, as long as you have completed a specialisation year at secondary school (at least one extra year).
There are no quotas for places, except for the artsy degrees. The year below me in my subject was so large that students were sitting on the stairs in the auditoria.
The upside is that what you decide to do at 12 or 15 doesn't impact what you can do later on. The downside is that the pass rate in first year of university is less than 50% (somtimes substantially so) for all subjects except medicine.

In Ireland you choose your courses and list them in order of preference online. Then when the results come out in August everyone's points are calculated (best 6 subjects) with certain grades conferring certain points. Then everyone who applied to a course is listed from highest to lowest points and those with the most points offered the places first. Most courses do also have requirements (say a certain grade in Maths) so even if you had higher points, if you don't make that requirement you won't get the place.
A bit like school admissions, if you qualify for more than one place, you are given the higher one on your preference list and the other place is filled by the next person on the waiting list.
The newspapers publish a list of the points required for each subject and university. Obviously the points go up and down depending on demand. It seems brutal, but fair and transparent.
No predicted grades necessary for either.

We used post exam grades to qualify for university. A straight point system which has a cut off depending on how many apply. You get the marks above the cut off you’re in. No interview, no predictions.

I know it won’t change here but man it’s full on with it all here.

Yes Gwyn they are. Which is why I'm probably not fighting harder and he actually isn't that bothered, it's me sad that has been riled by all of this.

@Aragog I’m guessing your a teacher. If a teen is producing B grade work with some A grade - say 60:40, I’m guessing you’d give a UCAS prediction of A. It’s a possibility but not solid. Whereas you’d give a ‘working at’ grade of B. What if it then goes to CAGs - does that teen get the possible A or the solid B?

Your = you’re! 🙄

Arog, in the situation you describe, if a school has to give CAGs, the grade given will depend on precise instructions from gov.

This year, schools were told to rank students and bear in mind historical averages. So, if 20% historically got A and the boy in question wasn’t on the top 20%, quite likely his CAG wouldn’t be A and then further down the ranking, the less likely school would give an A because they would expect Ofqal to adjust grades to reflect historic average. So this year even if school ignored historic average and gave an A Ofqal would have probably downgraded boy on 13 Aug. However with the way things worked out, in the end the original CAG would have been reinstated.

So this year, schools who ignored historic averages and wildly wildly raised CAGS benefitted most in the end. Those who actually stuck rigidly to instructions will have seen little change in results from previous years.

A CAG wasn’t just about an individual student but also determined by where they were in the rank order. If teachers felt others were more likely to get the top grade and had exceeded their ‘quota’ then some wouldn’t get the top grade.

When exams are set the same thing happens...sort of. Perhaps 10 are predicted A and 5 achieve it. School knows probably 4 or 5 will get it, but don’t know which 4 or 5 from 10, so predict all the A and then student exam performance determines the actual outcome. (In actual fact about 17% of ucas predictions are accurate). With the CAG system the school has to choose the 5 most likely rather than exam performance determining it. The trouble is that all 10 feel sure they would have got it. They wouldn’t all have achieved it but they All think they would because they had the ucas prediction and think they would have worked hard and done well in the exam. That’s where the disappointment lies.

It’s why Gavin is today stressing exams must happen in the summer. Only when students sit the exams do the grades feel earned and fair. Lots who sit the exam won’t achieve their ucas prediction and be disappointed but because they say the exam it will feel they earned their grade themselves rather than it being given to them to fit an algorithm or because of ranking that was beyond their control.

If exams are sat, exam boards can still decide exactly how many get each grade by adjusting boundaries after raw marks are in. This is much tge same as the Ranking process but just based on marks earned in exams not teacher ranking and based on a whole national ranking of marks not just one school. What might have earned an A* in one year might not earn it in every year. But it doesn’t cause upset as individual exam performance seems to determine the grade to students and parents. In reality, relative performance has always been important.

Sorry....but long, but I think we will see exams this coming year as it will avoid them troubles if this year.

Exams should never have been cancelled this year. It's led to absolute chaos in HE which will adversely impact this years cohort. The HE system was already in need of reform but this year has shown all its flaws and deformities. The current year 13 will have to now deal with the debacle of this year and pay the price. It's absolutely diabolical that there's not more being done for them. 3 weeks into term and they're being told there's likely to be more school closures, the curriculum could still change, the exam dates could change, every piece of work done could be used for cags..... it's scandalous.

In answer to your original question @Aragog I think both are needed and both need to be stressed to kids & parents. However, if a kid is working at high B, low A they’d get A aspirational and B target. But as time goes on that target may move to an A, a solid A. So the aspirational is no longer that aspirational but relatively solid (as solid as anything can be relying on a few exams). But yes, parameters would be useful.