I'm just going to quickly comment on the ASCL document that provides "Guidance regarding centre-assessed grades for summer 2020" found here: https://www.ascl.org.uk/ASCL/media/ASCL/Help%20and%20advice/Leadership%20and%20governance/Coronavirus-Guidance-regarding-centre-assessed-grades-for-summer-2020.pdf

The timeline part is fine, as it says it is indicative, and although I don't agree with all the elements... it is only indicative.

"Don't rush in to the process". Agree with this.

I think this "this is essential" part is a bit contentious. What has happened is obviously out of control but really we need to trust schools within a reasonable tolerance of their previous cohorts. This is desirable at a national level, but essential at a school level?

Yep, fine. Ultimate responsibility rests with the centre. But remember, teachers and subject leads are the ones with the best knowledge of students in their area.

Yeah OK this makes sense.

Right, I think the language here is too strong. Students may or may not need an 'objective mark' in a subject. Some subjects lend themselves better to this approach than others. I don't think it is vital but it could be a factor to leave to the discretion of the subject leader.

Second point on this, yeah we want to avoid unconscious bias, but the creation of an objective mark, when that is not normally done may in fact create some bias. What I want to say is that subjects need to do what they are happy with...

if the centre feels it needs to be more robust, as in more objectively created then that is a conversation to be had but it is up to individual schools and teachers to make those decisions which could be based on a whole host of localised knowledge, experience etc.

Yeah, fine.

I think the boys/girls point especially would just be borne out naturally in the data centres (teachers) provide. I don't think it needs special mention because isn't that just introducing unconscious biases that the document was keen to avoid earlier?

I think all of this is up to centres really. The special consideration part, I don't know really, it depends.

OK so, we're talking about after the grading and ranking process sanity checks here - I think, I don't like the word appropriate really.

OK so transition matrices could be used as a sanity check. I don't think they 'need' to be used. It does make me worry that schools may think that teachers need to be looking at transition matrices when making decisions about grades and ranks.

Again I think doing that would again introduce some biases into the thinking. I don't think this is what the guide is saying to do but it is open to (mis)interpretation.

I think this is dangerous for interpretation. Sure it is possible, but these are pretty big assumptions, especially for smaller cohorts. Secondly, a value added of around 0 does not mean you follow the national distribution, it could be achieved without the national distribution.

FLAW of the transition matrices, not the ASCL document is that they use sub-levels. Not many schools deal in sub-levels any more. Although they are easy to calculate, it is a way of looking at the data that is no longer familiar.

I'm not sure we need more unfamiliar things adding to an already unfamiliar process. Especially if this is being considered at any point other than a centre-level sanity check.

It's just too many numbers and may not represent the cohort in front of you. I know it's an Ofqual thing but the U grades make me sad. "I know you didn't have a chance to prove this wrong yet, but here's a U grade anyway".

Exam boards should offer free resits to all students awarded a U grade via their statistical process.

I think this is the biggest "if" in the document. Such a big "if" and if you consider yourself out of line, how much to adjust by? So many variables.

I had to break off to make dinner. Carbonara, for those who are interested. Anyway I was almost finished...

What ASCL have written makes some sense, if the sole purpose of this was to reproduce the national distribution from previous year... this part makes schools think that way.

However, surely the main purpose of this exercise is to award grades to students that is as fair a representation as feasibly possible of what the may have achieved. Now I do understand that we don't wish for national grade distributions to get out of hand but...

I think we need to start from a basis of trust and none one of statistical supposition. Once the full national picture is known it should be left to those 'behind the curtain' to determine the viability of the nations results and to make sensible adjustments accordingly.

That's what this part of the guide is saying... and I think as long as centres submit sensible data then this is where it should remain because there will be greater between-school inconsistencies if schools try to do this themselves...

and this will detract from what I think is the main aim, which is serve the students denied their opportunity (at this time) so that they can move forward with their lives.

And I don't think we should lose our vision by creating a data fog that confuses the fact that teachers know their students better than subject leads, who collectively know better than the centre, whilst centres undoubtedly know their cohorts better than a national statistician.

And I applaud ASCL for making a document that is intended to help schools at this difficult time but I feel in parts in veers from suggestion to instruction. I also think that the instruction parts sends schools down paths they may not wish to tread, looking at unfamiliar things,

that they do not need to look at it. This is clearly an unprecedented and difficult situation, schools know what they need to provide but I feel they should remain free to make that provision in a way that they are comfortable with.

At the end of the day, schools have to front some of these choices, with for example parents, who will have children in other years, they need to be happy with their methodology and to be making it for the right reasons, their reasons, right for them and their communities.