Most liked posts in thread: Exam results

They are the stats that have been being shown on the News so I don't have a link I can provide. But it was being used by BBC, ITV and Sky and I would assume others

A quick web search did bring up

https://www.ucu.org.uk/media/8409/Predicted-grades-accuracy-and-impact-Dec-16/pdf/Predicted_grades_report_Dec2016.pdf

Which says in the conclusion

I find a high level of inaccuracy of grade prediction. Among the best 3 A-levels students achieve, only 16% of higher education applicants’ grades are accurately predicted. However, the vast majority of applicants actually receive predictions that are too optimistic for the grades they actually go on to achieve, with 75% of applicants achieving lower grades than predicted

So 16% are correct with 75% getting lower than predicted.

I can't find the margin of error ones so you will just have to go back and watch old news reports if you really want to see them as I have better things to do than trawl through the Google rabbit hole.

And I don't know what is the best way but the simple fact of the matter is the algorithm resulted in schools performances being more in line with previous years than the predicted grades have. Schools grades don't fluctuate that much in reality, certainly not over a 5 year period that was being used so it seems about as accurate as you can get. And schools had the opportunity to appeal across the board under the original idea so if they could show why the algorithm was unfair to them for example maybe they can show a consistent improvement year on year at their school and they could get them adjusted.

I appreciate in some cases people would have got screwed over by the algorithm but some are also getting screwed over now with predicted grades, infact you are likely to now see some people have their grades fall. Giving many grades they wouldn't have achieved though isn't a solution

 

I'm assuming you went to Uni. You do remember your first semester right, hey the first year tbh. It's not exactly that intensive. I was in lectures maybe 3 or 4 hours a day with 1 day off completely for much of that first year, the hardest thing about it tbh was getting used to the day I had a 2 hour lecture at 9am and then not having another lecture until 7pm. Getting used to doing classes after 4pm was tough in itself but balancing that wasteland between the two lectures so you were in the right mindset for the later one was tough.

It's not like they need classroom time as long as its done within the first semester of uni (which if Uni's are open there is no reason why exams can't be taken, they can use the uni's examination spaces). The biggest issue is when the exams are taken, if its during the week it needs some flexibility from both the unis and exam boards as either the uni needs to shift lectures or every exam is going to need multiple different exams of the same level so that you can book a time to take it around your lectures. Although they could just do them on Weekends

And its not like they would have to take it. In general your A-levels are less important when you have a degree so it would be upto each student. Do you want formal A-level grades or have a need for them OR is do you trust in your ability to get your degree and that that will be more important in later life

 

Where are you getting they will be getting an aspiring target and one that they are currently working at? What they are actually getting is their 'predicted' grade which is full of 'bias' despite the predicted grade they are using being the one after the mock exams so in theory has been adjusted and the adjusted grade that tries to take account of that. They then get to choose which ever is highest.

If the grades were unbiased we wouldn't be seeing an increase that is almost ten times the normal year on year increase. Even with the adjustments its a bigger increase than we have seen year on year for atleast the last decade.

 

Where did I say anything about hard work? If you think your kid working hard has anything to do with how well they do in comparison to others then you are massively deluded. You can have the most dedicated kid in the world who spends 5 hours a night revising and doing independent study and they will still most likely be beaten by someone who doesn't try in class, would rather be down the park either playing football or drinking etc etc. Working hard might edge you over to the next grade but it absolutely does not guarantee you do better than those who try less hard.

If teachers could be trusted to get the predictions right we wouldn't have standardised testing still due to the negatives it brings. But they can't, Uni's have also been complaining for years about the fact they have to make decisions on the inaccurate teachers predictions. Really we shouldn't be relying on just one exam, strangely when I was doing A-Levels certain subjects were actual modular, for example in Business Studies we did an exam after each module but in Computer Science and Maths and the AS Accounting I did in the second year it was just a single exam. With the move to the A2 system it seemed that even more they were going that way but I believe its pretty much been reverted back to a single exam and even in many cases not even having a coursework element. If exams were being taken after every module it would be much easier in situations like this for teachers to give a fair mark as they would already have a fair chunk of their final mark in the bag and you could extrapolate from that.

 

The thing is though the prediction that will go down officially won't be the 'this is what you are working at' grade, its the 'this is what we think you can do' grade. If it was the you are working at grade then it wouldn't only have a 16% accuracy rate or at the very least if it was still 16% it would be 78% achieved 1 or more grades better than the prediction not 78% performing 1 or more grades worse than the prediction. Again what you are talking about is teachers attempts to motivate students, especially students who don't have an expectation of carrying their education on so need that push to better their chances in life

 

Only 10% get a grade or more higher than they were predicted. 78% get a grade or more lower. So no it doesn't even out

 

As I've mentioned an official study done for the Universities found only 16% were accurate. 78% were over inflated by 1 or more grade. That's a pretty big difference don't you think.

And Unis have been moaning about predicted grades for years, so much so that before the pandemic hit they were seriously considering scrapping the process of applying BEFORE they had their grades and instead handling the whole process between getting the results and Uni starting

https://www.bbc.co.uk/news/education-51645826

Yet once again studies have consistently found predicted grades at both A-Level and GCSE to be consistently over inflated. He may not be intentionally signing off on over inflated predictions but that is what is happening.

And there was a provision by the way for schools like his. Schools could appeal and get them adjusted if they could show why the 5 year history being used was unfair. If there has been a marked improvement in the school there would be data to show that and he would have had a case to get the whole school adjusted

How is it the right decision. Even via the algorithm grades had risen by around 1% above the normal rise. With going to teachers predicted grades we are looking at figures that will show almost 10 times increase on normal years. That just makes a mockery of the whole situation.

Yes there will be some who via the algorithm fell in a situation where it is screwing them over because without exams you can't account for the outliers but that's the same in any system that the outliers get screwed over. The 10% who outperform their predicted grades even in the algorithm version were getting screwed and most likely now find themselves getting screwed more (I bet if we took last years A-Levels and did them on predicted grades not only would that 10% in a worse percentile on the algorithm than they archived but going on predicted grades only I image that percentile gets even worse