Read on Twitter
Feel like another #ResultsDay #ALevelResults rant. Some people are arguing that it would be unfair to award CAGs since teachers must clearly have been overly optimistic in order for the CAG grade distribution to be so much higher than previous years', "necessitating" moderation.
But what are CAGs? They're supposed to be teachers' best assessment of "the most likely grade a student would have achieved if they had sat their exams". Imagine all teachers had done this absolutely perfectly. Would the CAG distribution look like a usual exam grade distribution?
Of course not. You would predict a distribution of perfectly assessed "most likely" grades to have more of the top grades, for example. A student may be "most likely" to obtain an A*, for example, yet actually have a bad day on the day of the exam, dropping down to an A or B.
For the other grades, movement is possible in both directions, but the probability distribution is not symmetric. A particular student "most likely" to get an A may, say, achieve an A* with 10% probability, an A with 60% probability, a B with 25% probability, and less with 5%.
If you add up students' "most likely" results you are not going to get the same as if you added up "real" results where you, over the population of all candidates, see the less likely results actually occur. But arguably the "most likely" grade is the one the student "deserves".
So it's not really the distribution of most likely grades that's necessarily unusually high, it's that in a "normal" year, many students miss out on the grade they deserve because they have a bad day (while a few get lucky and the thing they crammed the night before comes up).
Which means that our usual assessment process is unfair from this point of view, and more holistically obtained teacher-assessed grades can potentially actually provide a better assessment of a student's ability and the grade they deserve (though nothing is perfect).
Ofqual obviously realises all this, but also kind of doesn't... Here's how they phrase it. They're missing the point that if teachers had taken students' "bad days" into account, they would have disregarded the instruction to provide the "most likely" result.
Were they supposed to use a RNG to pick who'd have a "bad day"? Use a completely arbitrary, subjective assessment of how likely different candidates would be to suffer from nerves during exams...?! @ofqual
I should say that many centres will have tried to make their CAGs align with historical (exam-based) results, and so may actually have been a bit pessimistic with the CAGs (as described above, a perfect "most likely" grade distribution should be biased to have more top grades).
Anyone from a very small cohort was awarded their CAG without any scrutiny, and those from slightly bigger cohorts had at least some weighting from the CAG. This apparently led to the rise in top grades that students from larger cohorts with no CAG input did not benefit from.
Given that some students were awarded CAGs without scrutiny the only fair solution at this stage seems to be for all students to be awarded their CAGs. Yes, some may unfairly benefit, but at least those unfairly benefiting will not be limited to those from small cohorts!
