Thread by @ChrisGiles_

Ofqual has published guidelines on how it might revise grades based on “valid” mock results

Everyone knows this to be a bit of a panic response to the disastrous moderation exercise

Also entirely unnecessary if brains had been engaged earlier

1/
gov.uk/government/new…

The twin disasters of the process were an inability to allow for exceptions - the star pupils in ordinary schools (marked down) and occasional horror in a top school (lucky sod) - and decision not to standardise everyone, providing systematic bias in favour of private schools...

... with small cohorts

The peculiar achievement was to exacerbate unfairness with a standardisation process normally designed to encourage it. 👏👏👏

While we know no process can be entirely fair, it is easy to think of a fairer process - one that’s already essentially used in GCSEs and KS1 & KS2 SATs at age 7 and 11

The principle is simple.
1) Believe teacher grades
2) unless there is evidence not to and,
3) if so, apply a uniform adjustment using the best evidence you’ve got to the whole school’s results

This puts the pressure on schools not to inflate grades and each subject will monitor others to ensure the whole school doesn’t get moderated down

It allows outliers to get the top grades in previously low performing schools, doesn’t advantage people studying unpopular subjects or private schools

Clearly there could be gaming (eg schools with an A*A*A*A* pupil decided to allocate A*A*A*D to allow other pupils higher grades, but this could be dealt with with crude rules such as only the top 3 A levels count

And the would be arguments about the fair basis for a school’s expected value added given prior performance (but those exist anyway)

It would not have provided the same distribution of grades as in the past, but that’s a small price to pay compared with what DfE chose, which introduced specific biases and failed to allow moderated teacher assessments to give the top grades in many distressing cases