SET standardisation for age

[phantomguzzler]

Does anyone know for sure how they standardize for age in the first stage? I had been led to believe from this message board that they compared children born in the same month which seems the best way to me but now I have my doubts and think a pre defined number of points are added instead which could lead to different results. Before I check with the schools wondering if anyone else has checked and knows for sure how it is one. The schools seem to just say things such as "an age adjustment is made" or "a formula applied".

[streathammum]

As far as I can work out, there are different ways of standardising for age - one is where each birth month is standardised individually (I think this happens in Bucks and maybe Kent?), but there are other ways which involve applying some sort of standard formula across the results as a whole to weight for age difference.

My impression from looking at this last year was that the second stage tests at Sutton used this second method because the cohort (less than 650-700 overall, so less than 60 for each birth month) is too small do to standardisation by month. Not sure about the first stage, but even that would have quite a small cohort per month of around 500 children.

The schools should be able to tell you what they do. I would be happy to be corrected if this is wrong.

[surmum]

i am not too sure about the standerdisation. My DD sat for sutton test and did not clear, however i feel that this might be because of her birth date which is in SEPT as i heard that the standerdisation is for the standard formula accross the results as a whole to weight for age difference and not as per the birth months. Is anyone aware if we can appeal against this?

[BlueBerry]

I sincerely hope that my post doesn't offend anyone. I've been reading this thread and I'm getting a tad confused.

Why would anyone want to appeal the decision for the first round? Why would you want to put your child through the stress of scraping through on an appeal to sit the second round when there are hundreds of children who passed outright? I thought that most of the Sutton schools operated on rankings?

I really don't wish to upset anyone, as this is an emotive time for us all, but am I missing something obvious here?

[streathammum]

You can't "appeal" the results of the first round as such. Appeals are only possible after offer day (1 March 2018) - you can only appeal for a school you name on your CAF but aren't offered on offer day.

I think some schools do allow parents to request a re-mark of papers (for a fee) but the Sutton schools don't offer this. The first round exam is multiple choice marked by machine so I guess mistakes are unlikely.

In this particular case, there was a problem with the exam paper in the first round exam and this may have caused some children some difficulty in answering. The school has therefore checked all papers by hand and standardised them to account for this problem. I think the OP was wondering whether it's possible to ask the school to check this standardisation process again - the school's answer has been no.

The thing to bear in mind here is that there will be plenty of children who are invited to the stage 2 test *and pass* but who still aren't offered a place at these schools. More children pass stage 2 than there are places available. So a child who doesn't clear the stage 1 hurdle will find it difficult to argue that they're more deserving of a place than a child who did pass stage 2 but not well enough to gain a place.

[scary mum]

i am not too sure about the standerdisation. My DD sat for sutton test and did not clear, however i feel that this might be because of her birth date which is in SEPT as i heard that the standerdisation is for the standard formula accross the results as a whole to weight for age difference and not as per the birth months. Is anyone aware if we can appeal against this?

Standardisation is designed to make sure the system is fair, so I wouldn't worry about that side of it.

[Ladymuck]

Well, for Wilsons, the SET result only contributes 20% of the final score for ranking. So if I have a child who didn't manage to be speedy enough to finish one of the mc papers, but I knew would do well in the format of the 2nd round, I would be disappointed that they didn't get a chance to demonstrate their ability. Wilsons seem to indicate that rankings can change quite a bit from SET to final outcome.

Certainly some of the Sutton grammar schools used to add on additional marks (either 0, 1, 2 or 3) depending on which month you were born in, with a maximum of 3% being added to a summer born child. The adjustment isn't huge.

[phantomguzzler]

I am only interested in understanding the process not in appeals. I was led to believe time and time again by posters on this message board that each month was standardized which seemed the fairest way to me. To now find out that that might not have been the case and a measly 3% added to summer born children. What I want to know from the mathematicians at the grammar schools that the same children would have passed either doing the standardization the points added way or by comparing children and standardizing by month. I would like reassurance that proportionally the same number of children succeed for each month. For example if there were 300 September born then the top say 100 got through and if there were 150 August born children then that would mean 50 got through roughly to second stage (assuming top third in each month). That is all I am digging for, am not interested in appealing but am interested in a fair process. If indeed a different result is reached by adding 3% such that instead of the 50 August children passing only 12 do then I think that is not as fair of a method of standardization. Hopefully the same result is reached both ways.

[Ladymuck]

Posters usually point to the GL/NfER explanation of standardisation in order to give an idea as to how the process works, but each test is age-adjusted according to how that particular test-provider operates. Sutton have never, or at least not in recent memory, used GL tests, and typically simply wouldn't have a large enough population to comfortably standardise month by month. The adjustments that I have seen in operation have been done in 3 month intervals, so Sep-Nov, Dec-Feb etc.

[Bluearrow]

I would like reassurance that proportionally the same number of children succeed for each month. For example if there were 300 September born then the top say 100 got through and if there were 150 August born children then that would mean 50 got through roughly to second stage (assuming top third in each month). That is all I am digging for, am not interested in appealing but am interested in a fair process. If indeed a different result is reached by adding 3% such that instead of the 50 August children passing only 12 do then I think that is not as fair of a method of standardization. Hopefully the same result is reached both ways.

The old KS2 SATs marking scheme usually has a table which converts the raw score to standardised scores for children born in different months. For example, a raw score of 75/100 means 110 for children aged 11Y10M, and the same score is equivalent to 115 for someone who is 11Y0M. The reason is probably based on analysis of a large sample of children.

As for the grammar schools in Sutton, they do not make the information public. So no one knows exactly how it works. Due to the small number of students (even at a few thousand is still low comparing to the students of SATs), I am more convinced that they will standardise the score based on the months. Maybe it is not for each single month, but each quarter instead.

No method is going to be 100% fair to all children. But I believe they are trying to make it fair to most of children.