Confusing scores

Eleven Plus (11+) in Kent

Moderators: Section Moderators, Forum Moderators

Post Reply
11 Plus Platform - Online Practice Makes Perfect - Try Now
Twinkle
Posts: 580
Joined: Tue May 05, 2009 10:26 pm

Confusing scores

Post by Twinkle »

Our DS sat a 'mock' 11+ Maths paper ( Nfer) at school recently. We were told that his raw score was 42/45 and that this standardised at 125 ( out of the possible 140). He has just turned 10 - in other words he is right in the middle of the age range. We are flabbergasted that a score of about 93% as a raw score can translate into such a low standardised score. Can this be right? We were under the impression that a raw score of 70 - 80% in any of the Kent papers translated into a score of 140 after standardisation depending on the child's age ( in another thread ( "Odds of getting Judd, Skinners or TWGBS for their DS", SSM and tired_dad_2008 stated that 80% for Maths last year was sufficient to get 140 , following on from comments made by the HT at Judd) . Our DS is very good at maths and we were surprised he slipped up on the questions he got wrong,however we wouldn't have thought that getting three questions wrong would effectively put him out of the running for Judd or Skinners ( and thats even before taking into account VR and NVR scores). Can anyone help me understand this? Is it possible that this nfer paper was very easy and therefore in order to get 140 one had to get about 100% raw score? Seems odd to me. Btw we have actually had sight of the paper and the raw score is correct.

When we queried the position at school we were told it was an "odd paper" and dropping even one raw mark had a huge impact....

Bizarrely DS sat CATS papers at the same time and got 140 in the maths paper. Whilst I appreciate the papers will contain different questions there is a huge difference between 125 and 140 (1 standard deviation, or approx 91st and 99th centile respectively, which I'd regard as an unacceptable error range for an experiment).

Is it likely that the school is not sufficiently statistically literate to apply the tools that they've bought in correctly? Or are the tests measuring very different things?

As I understand it the CATS raw and standardised scores are marked and calculated externally whereas the NFER marks are based on a tool that the school buys in and uses itself.

How can I establish where DS is in the cohort and what, if any, he should pay attention to? He is predicted one level 5b and two level 5cs at the end of Year 5 on QCA testing.

Finally all the information that I have about the 11+ derives from playground chat and from this site ( which is great!). However does anyone know whether any official information about the substance of the Kent 11+ exams / scoring of the Kent exams is publicly available? I can't find anything on the KCC website.

Thanks

Twinkle
SSM
Posts: 646
Joined: Sun Feb 08, 2009 12:09 pm

Post by SSM »

As the school have already said it was an 'odd' paper, I wouldn't worry about this too much. The paper might have been very easy, thus a lot of children getting a high raw score.

As far as I know, there is no formal information published about the Kent 11+ other than it is Nfer MC. They even tend to change the timing format of the NVR, which is done in sections.
sp
Posts: 379
Joined: Thu Mar 09, 2006 7:09 pm
Location: groombridge, e.sussex

Post by sp »

Doesn't look like it's been "standardised" at all to me!! They've just reworked the original score to give a mark out of 140! Maybe they don't understand the real system. Like you, having read previous threads re raw and standardised scores, the mark looks brilliant so don't worry.
Bexley Mum 2
Posts: 851
Joined: Sat Nov 17, 2007 9:55 pm
Location: Bexley

Post by Bexley Mum 2 »

Twinkle - I believe one of the Nfer maths papers is quite a lot easier than the others (paper C I think). It may be that if lots of the kids in your son's class found it easier, when the school applied their standardisation formula it made it hard to get a high score. This won't happen when standardisation is applied across the whole of the Kent cohort. I'm pretty sure from what I've read on other threads that a score of 42/45 would give a standardised score of 140, or near enough, regardless of age. I presume if your child is practising 11+ papers at school they are at an independent. State schools don't practice for the 11+ (or aren't meant to!) so all those kids who haven't done any or much practice will impact on standardised scores.
perplexed
Posts: 490
Joined: Wed Jan 02, 2008 1:18 pm
Location: kent

Post by perplexed »

Standardising the results from one of those NFER practice papers is meaningless. Contact the publisher, NFER and Kent LEA and they will all say it is not a valid thing to do. I don't understand why the school would have even tried to give you the results in this way.

The NFER practice papers are just intended to show you all the types of question that could crop up. Each individual paper is not intended to be typical of a possible test paper. The school would not have the info on the whole cohort on which to perform a standardisation. Loads of the kids could already have done all the questions at home more than once. Complete and utter rubbish.

Sounds like he did extremely well.
Villagedad
Posts: 526
Joined: Tue Dec 23, 2008 5:22 pm
Location: Tonbridge & Tunbridge Wells

Post by Villagedad »

Does anyone know where you can get a free download of a sample 11+ Maths paper (Nfer) like they use in Kent.

Many thanks in advance
Villagedad
SunlampVexesEel
Posts: 1245
Joined: Fri Jul 06, 2007 9:31 pm

Re: Confusing scores

Post by SunlampVexesEel »

Twinkle wrote:We were told that his raw score was 42/45 and that this standardised at 125 ( out of the possible 140).
The test was clearly relatively easy however the standardisation helps to eliminate the fact the paper was either hard or easy.

The highest scorer will get 140 and the lowest 70 with 100 being the average.

Given only one score its impossible to see the distribution in this case but say... 39 correct was average and 44 correct was the best achieved in the test by all that sat it. Those getting 39/45 get 100; those on 44/45 get 140.

Your DS scores 125 which puts them well above average; in fact roughly in the top 10%. A good result.

See...

http://www.nfer.ac.uk/research-areas/as ... sation.cfm

for more detail.

Regards
SVE
Animis opibusque parati
Twinkle
Posts: 580
Joined: Tue May 05, 2009 10:26 pm

Confusing scores

Post by Twinkle »

Thank you all for your replies. Very helpful insights - I was tearing my hair out the night I posted. I feel alot more positive now.

This site is great ( I have lurked for a while) and is really my only source of information about the 11+. My DS is at a state school which doesn't support ( infact discourages) the 11+. It certainly doesn't coach for the 11+ and is very reluctant to provide any sensible information about how it works. The children were allowed to sit a practice paper in each subject so that parents could get an idea of whether their child was GS material / what school to aim for but in fact I think that the process has just led to confusion through what appears to be a very odd "standardisation" process.

I am yet to find out the raw score for NVR and VR so I may have more questions for the forum - please bear with me if the questions sound inane as I feel as if I am walking around in a sea of treacle.

Thanks again

Twinkle
SunlampVexesEel
Posts: 1245
Joined: Fri Jul 06, 2007 9:31 pm

Re: Confusing scores

Post by SunlampVexesEel »

Twinkle wrote:I am yet to find out the raw score for NVR and VR
The raw scores don't really help; they just give an indication of how easy or hard the tests were. The standardised scores are far more revealing.

Regards
SVE
Animis opibusque parati
Post Reply