Statistics can be manipulated to say what you like?

Eleven Plus (11+) in Birmingham, Walsall, Wolverhampton and Wrekin

Moderators: Section Moderators, Forum Moderators

11 Plus Mocks - Practise the real exam experience - Book Now
Dopey2012
Posts: 24
Joined: Sat Oct 13, 2012 10:46 pm

Statistics can be manipulated to say what you like?

Post by Dopey2012 »

Without knowing the precise statistical methods that are used to standardise scores how can parents make any rational decision. How do we know what difference increased or decreased number in the cohort will make to the scores? Or do those standardising the scores take those factors into account to try to achieve a comparable level from year to year. Is the younger age of the children taken into account? Would that raise or lower a child's mark and/or the lowest mark required to gain a place? Sorry for totally not understanding how to assess the chances.
MSD
Posts: 1731
Joined: Tue Dec 16, 2008 6:08 pm

Re: Statistics can be manipulated to say what you like?

Post by MSD »

Dopey2102,

The scores you received are age standardised. My son is a September born and achieved a score of 235 in KE consortium exam. If he was a June birth his score would have been 240+ for the same raw mark.

Whether I agree with it or not is a different story. I personally believe the children in the same year study the same curriculum, so it's unfair to penalize the older kids.

MSD
Okanagan
Posts: 1706
Joined: Mon Aug 22, 2011 9:20 pm
Location: Warwickshire

Re: Statistics can be manipulated to say what you like?

Post by Okanagan »

Dopey2012 wrote:Without knowing the precise statistical methods that are used to standardise scores how can parents make any rational decision. How do we know what difference increased or decreased number in the cohort will make to the scores? Or do those standardising the scores take those factors into account to try to achieve a comparable level from year to year. Is the younger age of the children taken into account? Would that raise or lower a child's mark and/or the lowest mark required to gain a place? Sorry for totally not understanding how to assess the chances.
Have a look here for an explanation of standardisation, and the age adjustments.
Dopey2012
Posts: 24
Joined: Sat Oct 13, 2012 10:46 pm

Re: Statistics can be manipulated to say what you like?

Post by Dopey2012 »

Thanks for both helpful replies. From what I can understand of those explanations it seems that scores gained in different years represent the same attainment level. If a larger cohort then naturally a higher score needed to gain entry? It doesn't seem to be the case that the same score in different years represents the same "rank". I still don't understand if the age standardisation is carried out relative to the whole cohort or absolutely in terms of actual age. If the latter then wouldn't all children need to gain higher scores this year with the exam being earlier? Am I overanalysing? How much might the historical pass marks vary by?
Priya01
Posts: 70
Joined: Mon Sep 10, 2012 10:03 am

Re: Statistics can be manipulated to say what you like?

Post by Priya01 »

I totally agree with MSD, the older kids are penalized for something they can't be held ownership for.they have put in their 100% effort like the other kids but can lose up to 5 marks just coz they were born couple of months earlier which is so UNFAIR.Anyway, there's no point stressing out now.just sit tight till march.
dr.watson
Posts: 153
Joined: Sat Mar 03, 2012 10:08 am

Re: Statistics can be manipulated to say what you like?

Post by dr.watson »

http://www.elevenplusexams.co.uk/advice ... xplanation" onclick="window.open(this.href);return false;
vza
Posts: 63
Joined: Fri Nov 18, 2011 10:35 pm

Re: Statistics can be manipulated to say what you like?

Post by vza »

MSD wrote "I personally believe the children in the same year study the same curriculum, so it's unfair to penalize the older kids."
Age Standardization does not penalized older children but tries to produce a level playing field for younger children.

The disadvantages of summer born children are well established and documented. It is well-known fact that summer born children do less well in school compared to winter born children and this trend continues into high school. It is not a valid argument to say, "Children in the same year study the same curriculum" so their abilities and development are not age dependent. If we simply consider vocabulary, which is a major part of CEM Test, winter born children have been exposed to English language almost one more year.
There are other factors such as brain development, problem solving skills and maturity, which are also age dependent. Summer born children are less mature and do not cope as well as older children with pressure, especially with CEM test with two papers of 45 minutes and 10 minutes break between papers.

VZA
UmSusu
Posts: 1015
Joined: Mon Feb 14, 2011 2:42 pm
Location: Birmingham

Re: Statistics can be manipulated to say what you like?

Post by UmSusu »

Completely agree with vza. It does sound unfair but if summer born children have been shown to consistently underperform in the exams then it is only fair to level things out a little - even then you cannot allow for the fact that there will generally be a difference in maturity. Leaving aside their actual educational achievement over the last year, surely your child has developed immensely in terms of maturity over the last 12 months?

Having had both a winter born child and a summer born child sit these exams I certainly noticed a difference in their ability to deal with the exam in a measured way. My summer born child is naturally much more intelligent and confident than my winter born child but was much more inconsistent under test conditions and in his approach in the lead up to the exams.

I did think it unfair in the past and felt the older children were penalised but if anything age standardisation only ever partly redresses the imbalance. It might well be true that in all their future public exams they will have to either measure up to older pupils or achieve less but in an an exam designed to test the potential of 10/11 year olds, surely they need to take this into account?


There you go : my kids have read a couple of my posts and joke that what I write is more like an essay than a post on a forum - I think I am beginning to see what they mean :D . Sorry for waffling on......

UmSusu
UmSusu
Dopey2012
Posts: 24
Joined: Sat Oct 13, 2012 10:46 pm

Re: Statistics can be manipulated to say what you like?

Post by Dopey2012 »

Interesting discussion but what I am trying to decipher is whether the age standardisation enables us to compare this years results with the historical results on a like for like basis because the whole cohort was three months younger this year...
Okanagan
Posts: 1706
Joined: Mon Aug 22, 2011 9:20 pm
Location: Warwickshire

Re: Statistics can be manipulated to say what you like?

Post by Okanagan »

Dopey2012 wrote:Interesting discussion but what I am trying to decipher is whether the age standardisation enables us to compare this years results with the historical results on a like for like basis because the whole cohort was three months younger this year...
In theory yes. But we don't know for certain if they actually recalculated the age standardisation tables of just applied a standard formula to the overall average to create them. My ds's primary say that since Warwickshire has had the CEM tests they've never had an Autumn born child get a place! Maybe just a statistical anomoly though - it's not a big enough sample to generalise on.
Post Reply
11 Plus Platform - Online Practice Makes Perfect - Try Now