To think that the Scottish government should not have caved over exam results?

[Notthemessiah]

So the Scottish government have caved in and have given their students the grades that their teachers have said they should get, despite the fact that overall they are massively inflated compared to previous years actual real results.

AIBU to think that this will massively penalise those pupils at schools where teachers were actually honest and realistic about their students likely results and instead benefit those who chose fantasy figures either through actual deceit, sheer wishful thinking or believing that the grades would be downgraded by some kind of system anyway.

Everyone was bleating about how it was unfair that pupils going to worse-performing schools got their results downgraded, but the stats don't lie - theirs were much more inflated compared to previous years that those from better-performing schools and it's ridiculous to think that all of them were suddenly going to improve this year.

It was always going to be an unfair system whatever happened but this just turns this year's results into a total joke - how universities, employers etc are expected to interpret them and compare them to other years is anyone's guess.

I hope that this doesn't happen in England and Wales too but it's hard to see how it can't - otherwise it puts Scottish students at a big advantage over their English and Welsh compatriots.

2pinkginsplease

I’m interested to know how you know 4 out of 5 grades were downgraded?

Fantasy grades are making a mockery of the whole process and teachers ability.

At my child’s school many of the teachers are sqa assessors so surely they know how to grade a child so work!

This year has been a year that children had to show resilience, courage, determination at doing work virtually and to keep up with their studies, surely our children deserve some credit for their hard work adapting to new ways of learning rather being put down and mocked!

@SmileEachDay I know because the head teacher phoned us on results day and told us that they would be appealing 4 subjects due to downgrading.

Dd received an ABBBC rather than her predicted AAAAB that she had been working at all year.

Predicted grades aren’t secret! And child can find ask their school for their predicted grades,

Yeah, I think you do need to have some skin in the game : or at least some knowledge of Scotland, its exam systems, and/or education in general to make any sort of substantive or worthwhile commentary on this. Otherwise, we risk spreading misinformation, stereotypes and generalisations.

The fantasy grades thing is insulting.

Re the comment about spot checks : I assume the problem is these didn't happen. That could have saved so much of this hassle and egg on face stuff.

2pinkginsplease

I see - I wasn’t sure how it was going to work this year. Predicted grades are, of course, not secret. CAGs on the other hand - I didn’t know they were going to be made public alongside the grade awarded to the child.

@Mistressiggi

Google something called the "burden of proof". The person who asserts something is the one who needs to prove it. It's like telling an atheist they have to prove God does not exist. Clearly you cannot.

And lol again - just knew that you would try the whole 'prove god exists' thing.

This is the definitiion from wikipedia

"The burden of proof (Latin: onus probandi, shortened from Onus probandi incumbit ei qui dicit, non ei qui negat) is the obligation on a party in a dispute to provide sufficient warrant for their position. "

Thing is we have two claims here:

1. I claim that some teachers have exaggerated some of their students grades well beyond anything they would ever have likely achieved in reality. Not all of them - just some. The figures regarding grade inflation, while not proving it, would certainly seem to support it.

1. You say that is wrong, so obviously your counter-claim is the opposite - that not a single teacher in Scotland (and now in England as the same thing is happening there) did so. Not one did so because they knew that the grades would be downmarked anyway, or because they particuarly liked one student and wanted to give them a hand towards that place they were after at the really good university, or that they simply made a mistake when filling out a form.

Now which of these is the by far and away the hardest to believe? I'd say it's you who now bears the burden of proof.

BTW - calling for proof in what is clearly a debate about opposing opinions is always such a lazy attempt at a 'gotcha'

they deliberately targeted children in deprived areas due to schools past performance, however individual children shouldn’t be graded on schools performance and on their own individual performance!

Jesus christ, they didn't target anyone - talk about using overly-emotive language.

Yes each child should be graded on their own performance (which is why we have exams) but they weren't possible this year so they had to pick a system by which to do so. The one they picked was to let teachers submit what they considered to be predicted grades (as well as, I believe, a ranking list of their students, at least in England). They would then use an algorthym to account for the fact that teachers were always, conciously or sub-consciously, going to inflate those grades.

This didn't target anyone and no more disadvantaged kids from schools in deprived areas than in afluent areas - anyone who worked hard and was likely to do better than their schools results for the past few years was at risk of not getting the grades they deserved.

Everyone was potentially affected by their schools past performance - not just those from deprived areas.

Here is an explanation fo how and why grades get 'inflated'

The grade a student gets in any exam isn't fixed until they take the exam. The grade partly depends on whether the questions suit the student and how they perform on the day. A good teacher has high expectations of their students and will predict the higher of 2 likely grades

@Notthemessiah I’m not using emotive language the stats speak the truth!

I’ve done my research on this topic as my child had been disadvantaged, the stats show that this has happened!

[quote 2pinkginsplease]@Notthemessiah I’m not using emotive language the stats speak the truth!

I’ve done my research on this topic as my child had been disadvantaged, the stats show that this has happened![/quote]
Then you aren't very good at interpreting them are you? All the stats show is that grades were more likely to be inflated in schools from more deprived areas then from others.

No targeting, no conspiracy, your child no more likely to be disadvantaged than any other by this stupid system (and I'm not doubting that children were disadvantaged or that yours was one of them).

I’ve done my research on this topic as my child had been disadvantaged, the stats show that this has happened

Your child was not “deliberately targeted” though - why would anyone do that?
It just means that your child’s school was predicting an improvement on previous years which seemed unlikely - not impossible, but unlikely.

The problem with using prelims for this type of exercise is that the SQA removed the facility for appeals a few years ago. You could only get a remark or a recount. This meant that prelims became a checkpoint rather than a formal evidence-gathering opportunity and so many schools changed the way they did them. The differences in the way prelims are carried out in different schools means they’re not really fit for the purpose they were used for this year.

My predictions are usually spot on. I normally get few kids who out-perform their predictions by a band or two and occasionally have some that don’t meet their prediction, of course. This year, I had one pupil who stayed where I put them and the rest were downgraded not just by a band but by a whole grade. I know for a fact and have evidence for the grades I put in. There were several rounds of moderation at department, school and local authority level. I don’t know if other teachers were over-predicting, but if they were it fucked us all over.

I’m glad for my pupils that Swinney made this announcement because my predictions were fair and evidence-based and it’s righted this wrong for them. I’m worried for my department for next year, though, and anticipate some kind of moderation-based stick being cooked up to beat us all with.

They are obviously expecting a massive shitstorm in England tomorrow with their "triple-lock" appeal and re-grade process they've announced.

All the stats show is that grades were more likely to be inflated in schools from more deprived areas then from others.

This is an interesting result and and it hasn’t been properly understood yet, although there are theories. Deliberate vindictiveness in targeting certain children seems implausible though.

Not RTFT because it is so long but yes I do think the Scottish government should have caved. An algorithm with a bias against poorer students is a bad thing.

And I really don't think that giving students a higher grade than they perhaps deserved really matters that much when their lives have been so badly affected by the pandemic. I am glad that they have introduced new rules in England and Wales too - it would have suited me just fine for A level as I got the same in my mocks as I did in my actual exams. I can't remember GCSE I might have done a bit worse but it's a safety net beyond which you can't fall, not what you get. For ds it would have been roughly the same, better in some subjects, worse in others and the same in others.

I never liked the idea of teacher assessment as there can be bias in that, too, but the system in place should work to address both that and the bias caused by statistical models.

Mind you, after all of this, they could have just let the teachers give the pupils their grades months ago!

Essentially, if every teacher decides those on a border will 'have a good day' so errs on the c rather than D side, as an example, some of those grades will stand, and some won't. Grade inflation occurs when they are all or many left to stand. But it wasn't any teacher putting in a fantasy grade.

Oh I like that Hamish. He's never shown any signs but I think I'll give him an A instead of a D is so not what happened.

An algorithm with a bias against poorer students is a bad thing.

It was not biased against poorer students. It evened out predictions which looked improbably high.

Why were there more of these in relatively poorer-performing schools? That’s a different question.

I really don't think that giving students a higher grade than they perhaps deserved

It matters to the students who were not given higher grades, because their schools were more cautious.

And what Piggy said. Taken individually each of these decisions looks reasonable, but the aggregated position does not.

Has anyone got a breakdown of results that has estimates versus actual for each grade?

The table shows the % of students overpredicted on the C/D going up as the schools get more deprived. I would predict the % of students overpredicted at the higher grades to go down as the schools get more deprived, and to go up for less deprived schools.

@kerrymucklowe2020

Why didn't they just grade the actual on the way they performed in the mocks- surely thats the fairest way for all?

It won't work that way fro all subjects or all schools. We try and make sure that the pupils take mocks seriously but they either do one of four things Take them seriously Take the core subject seriously and non core are after thoughts. Take those that they like seriously Take non of them seriously.

With that and some subjects trim the mocks so that they only have the components that they have done and others do an unedited various means that the results are just wild cards.

Also 40 to 50% of the marks for my subjects mean that I can't just use the mocks.

I would predict the % of students overpredicted at the higher grades to go down as the schools get more deprived, and to go up for less deprived schools.

... until you get to the schools where everyone gets high grades: (a) they’re much easier to predict and (b) there’s no upper borderline for an A* student!

I wonder if some of it is that disadvantaged students on average underperform in linear exams. The evidence tends to show them being stronger, but when it actually comes down to the exam, they miss it.

When we predict grades, we give students a fine grade:

Eg

4- we aren’t confident they’ll get it
4= We’re secure they’ll get it
4+ With a prevailing wind they might tip into a 5

Some of the 4- students will get a 4, some won’t - I’d imagine that’s the borderline that much of the inflation has happened.

(I know Scottish grades are a little different but I’d imagine there’s a similar process?)

Well , fortunately, the Scottish government has launched an independent enquiry so I am sure some interesting findings will emerge.

To be fair, any (slight) inflation of disadvantaged students was encouraged by both Ofqual and SQA who warned about unconscious bias and flagged the stats of historic underachievement (only to, whichever way you slice it!) embed it further themselves by focusing on schools prior performance!

The problem now is that those who get into university with these results will expect the next three or four years to be dumbed down for them. If they can't cope Covid will be a handy excuse for life.