Flight paths in secondary are nonsense and demotivating for pupils SAY OFSTED

Your last post, cool is despicable

I don't put any store in your value judgements, noble. Apart from finding it interesting that that comment is one that particularly riles you. Perhaps it hit a nerve? Do you pride yourself in being particularly compassionate towards parents?

No, you are applying cohort statistics to individuals.

The cohort results actually filter down and affect an individual's chances of achieving a particular result. Even if the adjustment is very small on an individual level it will make a difference and if that difference is on a boundary there will be a tangible effect.

Then you apply faux concern to a worried parent who doesn't fully understand her child's targets on their progress report. Which in fact is secret glee because the example serves your political and professional views.

You, noble, still haven't said what grading system you think would be preferable. I think you just want to hold absolute power over what assessment information and your interpretation of it is shared with parents, students and even your SLT and the government. You want to self govern, self assess and make unquestioned judgements concerning your student's attainment and progress. And then taunt parents with any snippet of information that suggests OSTED might just let teachers get on with it and not inspect very much of their teaching practices.

Ooh how many mathematicians do you know to know that most would do something monumentally stupid and ignore sample size?

Several who would explore the effects of sample sizes across different contexts.

Indeed it can, on a national scale, but not on an individual one.

That is a nonsensical statement. The cap is effectively applied across the board as it results in a lower actual probability of getting a high grade. As the overall grading is adjusted according to the KS2 results. This probability effects all who take the exam.

I don't put maths or mathematicians on pedestals either. And a lot, I know most definitely would give it a go just to look at the results. Because they are curious.

You can't take whole cohort descriptive statistics and use that to make predictions of individuals in a different population.

You can but accept this has limitations. Mathematically you could probably do some sort of Bayesian analysis...Anyway I diverge.

My suggestion is stop assigning grades to kids

This is where we differ. There needs to be some sort of grading system throughout if the aim is to achieve good grades at the end of the course. Otherwise there will be no consistent measure to ascertain whether a child is 'on track'.

Interesting, though, that if a cohort has lower KS2 scores (than previous years) then they will actually have lower GCSE scores, as a cohort (than previous years). Essentially KS2 can effectively artificially cap attainment. I can appreciate secondary school teachers will find that pretty galling. Dread to think what effect that might have in wanting to manipulate admissions.

I think it is this that needs addressing, though, rather than removing the need to give any advise regarding the amount of progression that can be expected.

Cool you don't seem to understand the statistics. You can't take whole cohort descriptive statistics and use that to make predictions of individuals in a different population.
That's one issue I have about your ideas.
The other is that even to do that you'd need national tests every year (every unit?) to analyse the data for the whole cohort. Single schools are not large enough samples. That's a complete overhaul of our education system.
My suggestion is stop assigning grades to kids!

So what is your alternative, noble?

Oh, I forgot to say apply the score adjustment.

So basically, the results in terms of grades are pre-determined, and the grade boundaries are set to ensure this happens.

As I mentioned, when I talked about 'deciding' the percentages of students achieving each grade. I acknowledged, also, political motivations affecting these decisions p. However every so often there is an examination of standards but nobody likes that as it usually means the system changes yet again which clouds comparisons further and creates more opacity within the system.

Now, how can you apply that to a Y8 end of year assessment?

By applying similar measures and extrapolating.

Look at the KS2 data. Compare it the KS2 data of the previous year's cohort sitting exams. Devise an adjustment based on this comparison. Then decide how many pupils should be getting which grade. Then produce a graph of how many pupils should be getting each grade (as if time had gone forward and they had taken the GCSE) and put that against the graph of actual test scores, and put a pin in each test score that will make the actual results best fit the (virtual GCSE level) grades the pupils are meant to be getting. From this devise the grade boundaries. Then look at the grades previous cohorts got in Y8 compared to what they got in the GCSEs. Work out the average progression. From that extrapolate down the grading from the virtual GCSE level to what is being achieved currently.

Because when you know how it’s actually done, that’s when you realise that your notion of giving grade boundaries in Y8 to a topic test are just nonsense.

Why don't you tell, me, then, noble if you want to attempt to prove your point.

The thing is, so far you have not actually offered any real alternative which even has any potential to communicate anything remotely intelligible to parents regarding their child's attainment and progress. You don't appear to believe the GCSE grades hold any validity, either.

So the education system becomes more and more like a mystery cult whereby only initiates are the holders of the 'secret knowledge'. Elitist, divisive and open to corruption. That might reflect the reality of the education system now. However, I for one, will never support practices which allow this continue.

I’m going to ask again cool because this is absolutely key.

Me revealing my own understanding on this really isn't 'key'. However, I will answer anyway.

I know that there are different schools of thought, in theory, regarding how do this. From setting boundaries based on the percentages of pupils from a particular year achieving raw scores so only the top x% of pupils raw scores can be assigned top grade and so on, filtering down. Because the top grade has to have an aspirational value. However this doesn't allow for consistency of standards between different years or guard against varying political motivations in having a certain proportion of people achieving various levels of qualifications affecting results. So then a wider analysis has to happen looking at setting grade boundaries related to the percentage of understanding of course content demonstrated.

So, yes, I do appreciate the complexity. But I'm all for exposing it rather than scrapping the system altogether. Because, like it or not, people have a need for classification. If you remove one sort another, not necessarily a superior one, will rapidly replace it.