Score standardisation

Eleven Plus (11+) in Buckinghamshire (Bucks)

Moderators: Section Moderators, Forum Moderators

bluesymum
Posts: 86
Joined: Mon Oct 10, 2016 9:33 pm

Score standardisation

Post by bluesymum »

Sorry if I sound a bit daft, I'm a bit confused about how the scores
Are standardised. It was suggested in a few of the previous
Posts that the score is influenced by the performance of the'cohort'
for the year. However, on the Bucks website, it states that the marks for
Each section are tallied and the total score is standardised
as per age and weights are given to each section(VR-50%, NR-30%
NVR-20%). So, the 121 pass mark is purely a weighted average of your
score? And not normalised( influenced by how others in the cohort have scored?)
:roll: :shock:
scary mum
Posts: 8841
Joined: Mon Mar 15, 2010 3:45 pm

Re: Score standardisation

Post by scary mum »

"Standardised by age" means that they are only compared to others of the same age. Therefore if, for example, the children born in March all score very highly, the raw score for 121 might be higher for that group. See the posts from yesterday where it is explained (several times).
scary mum
streathammum
Posts: 1252
Joined: Sat Oct 01, 2016 6:02 pm

Re: Score standardisation

Post by streathammum »

I've seen this written a couple of times now and I'm a bit confused. What you've described, where a clever month is lowered and a weaker month is raised, it doesn't match what's on this page about standardisation: http://www.elevenplusexams.co.uk/advice ... xplanation

The way you've described it, if October borns happened to get higher scores than September borns one year, the October children would be standardised to lower their scores slightly, to make them comparable with the September children. October born children would have to get higher scores than older children to get a pass mark. That doesn't sound right.

The link on this site suggests that standardisation only happens one way - to bring the later born children up to compare with the earlier born children - and that marks added at a steady, predictable rate month by month. And that makes more sense to me, because it assumes that the standardisation is only there to correct the slower development of younger children.
kenyancowgirl
Posts: 6738
Joined: Mon Oct 21, 2013 8:59 pm

Re: Score standardisation

Post by kenyancowgirl »

That might make more sense to you, but it is NOT what happens - and certainly not what happens in Bucks. Search for a recent post by MimiTa where there are several explanations about the same thing. Scary mum is absolutely right as to how standardisation works.
streathammum
Posts: 1252
Joined: Sat Oct 01, 2016 6:02 pm

Re: Score standardisation

Post by streathammum »

Yes, I saw all the posts yesterday and that's why I'm posting now.

Do you have a link for what happens in Bucks? If it is as you all describe, it's definitely not what is set out in the Advice section of this site (the link in my earlier post).
scary mum
Posts: 8841
Joined: Mon Mar 15, 2010 3:45 pm

Re: Score standardisation

Post by scary mum »

The way you've described it, if October borns happened to get higher scores than September borns one year, the October children would be standardised to lower their scores slightly, to make them comparable with the September children. October born children would have to get higher scores than older children to get a pass mark. That doesn't sound right
It's not to make them comparable with the September borns, it's to ensure that 30% (roughly) are selected from each month, overall allowing for the fact that generally, those older will tend to do slightly better due to having been on the planet for up to a year longer. Somewhere back in the dark ages on this section of the forum there are data from the old 11 plus showing raw scores and standardised score by birth month (search on author tree, standardisation in the Buckinghamshire section). If you can find it is would give you a sense of how it works, although it wouldn't be directly applicable as the 11 plus has changed, as have the scores given.
ETA:The way I read it, that's exactly what the section on this website says about standardisation.
scary mum
BucksBornNBred
Posts: 1031
Joined: Thu Aug 13, 2015 4:01 pm

Re: Score standardisation

Post by BucksBornNBred »

I have to agree with Streathammum. The explanation on the link does not appear to match the advice being given on the forum. Also, the two links at the bottom of that page don't seem to be working.

ETA: Reading that page again I think it might explain it, but I think the recent explanations on the forum make it clearer and lead to less confusion.
scary mum
Posts: 8841
Joined: Mon Mar 15, 2010 3:45 pm

Re: Score standardisation

Post by scary mum »

Another way of looking at it simplistically is that there are twelve buckets, one for each birth month. They are independent of each other and all the scores for each month are put into their appropriate bucket. Depending on what percentage you wish to have qualifying, the top eg 30% are creamed off from each bucket. The raw scores are almost irrelevant but may or may not be different for each month. I don't know if the statisticians would approve of this simplistic view, but this is how I think of it.
scary mum
BucksBornNBred
Posts: 1031
Joined: Thu Aug 13, 2015 4:01 pm

Re: Score standardisation

Post by BucksBornNBred »

I think if that link was updated to say that the top XX% are creamed off from each birth month then there would be less confusion each year. I know when I was new to the site I also read it as points being added/taken away.
streathammum
Posts: 1252
Joined: Sat Oct 01, 2016 6:02 pm

Re: Score standardisation

Post by streathammum »

I do understand what you're saying. It just seems to have the potential for significant unfairness.

I would expect age standardisation to bring up the scores of the younger children to be comparable with the older ones, based on an understanding of how, younger and older children of the same innate ability perform in the same test. I would not expect it to guarantee that there is an even spread of passes among children throughout the year, although often this will in fact be the outcome.

I would not expect that you would have a system whereby a younger child with a higher raw score could have a lower rank than an older child with a lower raw score. But the system you have described does allow for this.

It would happen if, in a particular year, you have a particularly able bunch of children who all happen to be born in March. In this case, the raw score you'd need to get for a pass as a March child would be higher than the raw score you'd need to get as a February child, even though the February children are older.
Post Reply
11 Plus Mocks - Practise the real exam experience - Book Now