How accurate are predicted GCSE grades?

[ifonly4]

DC is in Year 10 and started to get predicted GCSE grades. Just wondering how accurate they are, as they're not what I expected.

I've started another thread today that relates to this, but I can say that they are not accurate when it comes to my DD. I have been told and I vaguely remember reading the same here a year or so ago, that these predicted grades are based on a number of factors such as SATs results and socio-cultural indicators and these are set by the local authority.

Because my DD's primary school didn't do science SATs, the LA set her results for the purpose of predicted grades at the average 4b, meaning that her official Predicated grade is a B, that despite the fact that she got A* in the three mock exams she did last year, and her actual target grade.

I am trying to find out how this can be amended for the purpose of college applications.

You should not be given the computer generated target (not predicted) grades as they are not meant to be applied to individuals. Predicted grades should be set by the teacher and how accurate they are depends on how good the teacher is at predicting the future.

DS got much better results than his teachers predicted. In History he went from predicted E to actual A, and similar in French. And he didnt work in those two, assuming he would fail, he concentrated on other subjects.

I'm hoping ds3 will do similar next summer, as he's not doing too well currently.

When doing mine (A levels) I used previous test and exam results and the added/subtracted a grade or 2 depending on the attitude of the student - and stated this alongside the grade.

That way parents and students knew what the score really was.

But the machine generated ones are usually a mystery when applied to individuals, as pp have said, they are not appropriate.

Thanks for your replies. My DC's predicted grades are better than expected in most subjects, fantastic if they are achieved but don't want to feel pressurized/let down later on

From my experience of 1 child, at the start of year 10, I would say not very accurate. DD's were generally too high. As time goes on they should be adjusted and get more accurate.

DD got her final predictions Easter y11. Even then she was 1 grade lower on 3, 1 grade higher on 2, and exactly met 3.

However, I considered the 3 she didn't meet as over predictions and told her so (2 were just plain wrong, 1 was just slightly over-optimistic).

Pretty much spot on ... and two of my DSs got identical predictions and identical results.

You should not be given the computer generated target (not predicted) grades as they are not meant to be applied to individuals. Predicted grades should be set by the teacher and how accurate they are depends on how good the teacher is at predicting the future.

I think this is where the confusion is. It would appear that at DD's school, what they refer to as predicted is that computer generated target indeed, or at least this forms part of it. What you refer as predicted grades is what they call expected grades and that is what is set my teachers.

The whole thing is totally confusing!

And you also have to factor in whatever 'Progress' pressure the teacher, department and school may be under.

For example, in A levels, if the machine predicted grade was, say, C, but the student was pratting around and doing no work, or having a really hard time making the leap, I could not give an 'expected grade' lower than the C on any official paperwork. That would start a 'why aren't you teaching well enough for this student to achieve their predicted grade' inquisition. And no one wants the extra workload that entails, those 8th days in the week are on ration, apparently!

So your higher than expected grades could mean your DC is doing really well, their teacher has forgotten which one in the class they are, the teacher/dept/school is under pressure in that subject.

You can work it out by looking at any homework commentary, quiz/test marks and an honest chat with your DC. Then a quick chat with the teacher, who is more likely to qualify the details if you ask nicely and with some info to hand.

There is a difference between target grades and predicted grades.

A target may well be higher than the prediction! DS1's targets were all Cs but his predictions, certainly towards this time in yr 11, were a bit lower (and more realistic!).

Predicted grades at this stage of yr 10 tbh are a bit of a stab in the dark. I don't really know what my year 10s are going to get tho I can make a reasonably educated guess, I suppose. They may do better, they may do worse, depending how hard they work (among other factors). But I am basing it on levels in KS3 and my observation/assessment of them so far in KS4.

Throthickandthin if a teacher predicted an E in French and your DS got an A there was something seriously wrong. At what stage was an E predicted? By year 11 about 50% of coursework may well be done and surely a teacher can see if that is E or A. Students of mine who are producing E-level controlled assessment are rather unlikely to get an A tbh.

For Maths and English, predicting grades for Year 10 is virtually impossible (and will remain so next year). They are the first GCSEs to be using new specifications and 1-9 grades, teachers have few examples of 1-9 level work, the exam boards aren't proving mark boundaries. We're using FFT grades as 'targets' and then taking a best guess for 'predicted'. It really isn't going to be anything better than a guess for the next two years, until we've got past papers, exemplars etc to use.

We were given "targets" which DS would be capable of achieving if he worked. We were also give current grades, which were woeful much lower. He came through with mostly his target grades, just because he aced the final exams, despite not having put much effort into controlled assessments or coursework.
Teachers seem to have verbally given DS predicted grades, but we never had those reported to parents in writing.These were realistic, and in some cases he did much better, and did reach his target.

Targets seem to have been generated based on sats results and our postcode. DSs head of year in Y8 let on that DS was in the top 5% of the country or something, based on those figures. Sounded like a lot of nonsense to me.

Dd was predicted all B. She got 5's in the Sats but her.primary advised the school (grammar) that they felt she was more 4b

Pissed off as we were at this initially, it meant she didn't feel massively pressured and in fact got 2 A* 7 A and 5 B

However, she was also pretty.much un noticed throughout her 5 yrs there as was just average in their eyes. She got better results than most of her friendship group although they'd been predicted A's
She actually moved to another grammar for 6th form and they welcomed her at the booking in interview stating that she was an exceptional young lady. In all honesty, I'm not sure yet what's going to be better as she may not reach the grades they now think she can!
I think it's almost a 'how longs a piece of string' exercise

I'm going to ask at parents evening about how they assess "predicted" grades. I might be doing my DD an injustice, but I'd have thought she was likely to achieve As and Bs . So far has been predicted Chemistry A, Music A, Geography A*, Maths 8 & English 7. Chemistry teacher said he wasn't sure how accurate the grades are himself and DD is certainly at the lower end of those doing triple science. They've been done under Fischer something or other based on SATs level (achieved level 5 in everything), area (probably thought of as good) and income (we know lots of of others who earn £10-20,000 more than DH). I guess this has to be a general assessment as they can't possibly know what DH does for a living or earns.

My DD was given all D and E predicted grades. She worked hard and scored all grade C (except one D in History).

So I think they were fair predictions finally outweighed with hard work.

They've been done under Fischer something or other based on SATs level (achieved level 5 in everything), area (probably thought of as good) and income (we know lots of of others who earn £10-20,000 more than DH)

That's the one DD's school uses that ended up screwing the result. DD didn't do Science SATs, so was set as average (which she isn't), area probably isn't top, and same as income, hence her predicted levels being inaccurate.

Saying that, I raised it with the Year Leader who having spoken to her teacher last year agreed that they were set too low and will therefore be adjusting them from Bs to either As or A*s.

It really makes me cross that schools are misusing data in this way. Fischer data should never be shared with parents. It will only cause confusion.

They are average targets and will be roughly accurate over a cohort. Some people will do better than their FFT target, some worse and that is expected because the targets aren't intended for individuals. They can be used to tell if a year group has done better than expected or not, because that year group is a large enough sample.