Wrwickshire versus Bimingham
Moderators: Section Moderators, Forum Moderators
Wrwickshire versus Bimingham
I also received my daughter's test results from KE foundation office today (no i did not pick it up, it came through the post). I wasn't going to find out her result since we already got the first choice school (Stratford Girls Grammar), but there was a post in Warwickshire section about children having advantage by sitting the exam first hence familiarising themselves
with the format.
I was intrigued so i wrote to the foundation to request her results.
Her scores in the Warwickshire results were VR 143, NVR 143 and Numeric 139, giving a total of 425. In the Birmingham test, her scores were VR 147, NVR 145 and Numeric Reasoning 135, giving a total of 427.
Looking at her results, by sitting the Warwickshire exam first did not make a big difference. According to my daughter,in terms of the contents of the two exams , the format are very different.
I hope this can re-assure people.
with the format.
I was intrigued so i wrote to the foundation to request her results.
Her scores in the Warwickshire results were VR 143, NVR 143 and Numeric 139, giving a total of 425. In the Birmingham test, her scores were VR 147, NVR 145 and Numeric Reasoning 135, giving a total of 427.
Looking at her results, by sitting the Warwickshire exam first did not make a big difference. According to my daughter,in terms of the contents of the two exams , the format are very different.
I hope this can re-assure people.
Hi Concerto
Actually your daughters scores for the B/Ham test were very much better than the Warwickshire test.
Because historically the Warwickshire Test used to use Morey House which were Age Standardised against a representative National Sample rather than the cohort, Warwickshire decided to request that Univ Of Durham mark the new test in a similar way.
The results is the for all candidates the Warwickshire test results are always very much higher then normal with lots of candidates achieving close to the max scores in each section (somewhere in the 140s).
The results for B/Ham KE exams are quite different, these are Age Standardised against the cohort with the scores around 141 being well within the top 1%.
Your daughters scores for VR and NVR were probably very close to the top score in the B/Ham KE Test.
Very well done indeed.
Regards
Actually your daughters scores for the B/Ham test were very much better than the Warwickshire test.
Because historically the Warwickshire Test used to use Morey House which were Age Standardised against a representative National Sample rather than the cohort, Warwickshire decided to request that Univ Of Durham mark the new test in a similar way.
The results is the for all candidates the Warwickshire test results are always very much higher then normal with lots of candidates achieving close to the max scores in each section (somewhere in the 140s).
The results for B/Ham KE exams are quite different, these are Age Standardised against the cohort with the scores around 141 being well within the top 1%.
Your daughters scores for VR and NVR were probably very close to the top score in the B/Ham KE Test.
Very well done indeed.
Regards
Extremely well done Concerto and Daughter..
I can confirm that the VR and NVR were the top scores and with that excellent Numerical score, your DD without a doubt in my mind anyway, the top performing child in the Birmingham Exams.
This also now confirms what i believed all along that if you have a star pupil who is above the rest by a few points even after being standardised, can score well above the 140 mark.
My DS done extremely well and had a higher numerical mark. The 432 standardised was the top.
Well Done again and i am sure your DD will be a top performer in the years to come.
Quizzer
I can confirm that the VR and NVR were the top scores and with that excellent Numerical score, your DD without a doubt in my mind anyway, the top performing child in the Birmingham Exams.
This also now confirms what i believed all along that if you have a star pupil who is above the rest by a few points even after being standardised, can score well above the 140 mark.
My DS done extremely well and had a higher numerical mark. The 432 standardised was the top.
Well Done again and i am sure your DD will be a top performer in the years to come.
Quizzer
Wonderful results, Concerto, congratulations to your daughter! You must be very proud of her.
"Actually your daughters scores for the B/Ham test were very much better than the Warwickshire test."
I don't think it's as clear cut as that Ken, there are just not that many children as clever as Concerto's daughter - her score in both exams is about the 99.7 percentile, i.e. about 1 in 400 give or take (in other words, in most years a primary school won't have any pupil that clever - I hope your daughter's teachers appreciate her, Concerto!).
Off the top of my head, I think at her level the normal day-to-day variations in achievement will have more impact on her placing than differences in standardisation. We'd really need to see a comparison of people lower down the scale to get a clear picture of any differences there might be between the two tests.
Mike
"Actually your daughters scores for the B/Ham test were very much better than the Warwickshire test."
I don't think it's as clear cut as that Ken, there are just not that many children as clever as Concerto's daughter - her score in both exams is about the 99.7 percentile, i.e. about 1 in 400 give or take (in other words, in most years a primary school won't have any pupil that clever - I hope your daughter's teachers appreciate her, Concerto!).
Off the top of my head, I think at her level the normal day-to-day variations in achievement will have more impact on her placing than differences in standardisation. We'd really need to see a comparison of people lower down the scale to get a clear picture of any differences there might be between the two tests.
Mike
KenR wrote:
I am not sure what evidence there is to back up this statement. What I do know is that a daughter of a friend of mine achieved a score of 380 in the Warwickshire test and was ranked joint 17th overall. My daughter scored 316 and was ranked joint 369th. Overall in excess of 900 children sat the exam. This leads me to conclude that very few children would have achieved a score above 400. It seems to me that Concerto's daughter probably achieved a truly exceptional score." The results is the for all candidates the Warwickshire test results are always very much higher then normal with lots of candidates achieving close to the max scores in each section (somewhere in the 140s)."
369th position out of 900 would (in a statistically perfect world) be a score of 310, so for it to be a score of 316 is not impossible given that the candidate population was probably a bit brighter than the standardisation population. On the other hand, a score of 380 corresponds to about 32nd in an "average" population so 17th is rather better than you'd expect (unless there were quite a few children in 17th place of course!), but at the thin end of the results curve that sort of result isn't all that unusual.
In a test population of about 900 you'd expect around half a dozen to score over 400. [edit - if the candidate population matched the standardisation population. Perhaps more if the candidates are brighter on average - but not many more, as per my comment above - with only ~900 taking the exam there still won't be many children this high on the ability curve.)
Mike
In a test population of about 900 you'd expect around half a dozen to score over 400. [edit - if the candidate population matched the standardisation population. Perhaps more if the candidates are brighter on average - but not many more, as per my comment above - with only ~900 taking the exam there still won't be many children this high on the ability curve.)
Mike
There is no way I wish to devalue the the truly exceptional performance of concerto's daughter, but maybe I should explain my comments about Warwickshire a little further.
Warwickshire used to use Morey House papers which were similar to NFER type papers and already Age Standardised against a National Sample. All Warwickshire LEA had to do was feed the raw scores and ages into a computer program and it then it came out with an Age Standardarised score.
However because it was a National Age Standardisation, quite a lot of children achieved the top score (140) - this is because children who are entered for 11+ exams tend to be better performing.
For example, in the last year of the Morey House Tests for the Warwickshire Southern Area, 4% of candidates achieved scores between 278 - 280 and 23 out of 951 candidates (2.4%) achieved a perfect top score of 280. For a more usual Age Standardisation against the cohort less than 0.1% of candidates would achieve a top mark.
When Warwickshire switched to NFER last year and then Univ of Durham CEM this year, they wished the Age Standardised profile to be similar to previous years (so they coúld perform validation checks etc)
So what I'm saying is that although concerto's daughter achieved the top mark in Warwickshire, there might be maybe 2% of candidates that had similar scores.
Whereas in B/Ham, because the exam is Age Standardised against the cohort, we can be fairly confident that concerto's daughter was in the top 0.1% and probably the top 1st or 2nd performing candidate in the exam.
Hope this helps
Warwickshire used to use Morey House papers which were similar to NFER type papers and already Age Standardised against a National Sample. All Warwickshire LEA had to do was feed the raw scores and ages into a computer program and it then it came out with an Age Standardarised score.
However because it was a National Age Standardisation, quite a lot of children achieved the top score (140) - this is because children who are entered for 11+ exams tend to be better performing.
For example, in the last year of the Morey House Tests for the Warwickshire Southern Area, 4% of candidates achieved scores between 278 - 280 and 23 out of 951 candidates (2.4%) achieved a perfect top score of 280. For a more usual Age Standardisation against the cohort less than 0.1% of candidates would achieve a top mark.
When Warwickshire switched to NFER last year and then Univ of Durham CEM this year, they wished the Age Standardised profile to be similar to previous years (so they coúld perform validation checks etc)
So what I'm saying is that although concerto's daughter achieved the top mark in Warwickshire, there might be maybe 2% of candidates that had similar scores.
Whereas in B/Ham, because the exam is Age Standardised against the cohort, we can be fairly confident that concerto's daughter was in the top 0.1% and probably the top 1st or 2nd performing candidate in the exam.
Hope this helps
-
fm
I agree with Ken. Standardising against the general population is quite different to standardising within the cohort.
I also agree that Concerto's daughter is exceptional.
What would be really interesting, however, would be the raw scores which yielded these marks.
I have seen quite a few raw scores for the Sutton exams and the affect of standardising does distort the whole final score issue. For instance last year I had one candidate on 44/45 in VR who scored 141 standardised and one who had 43//45 which standardised to 133. One point could be down to little age disparity but basically the first child received a whopping 8 more, simply for getting 1 question more right.
Extrapolating from this, if the upper scores for KE trailed off at, let's say, 56/80 in the maths, but there were a fair few grouped round this score, they might end up with a standardised 130. Then you might get a child on 57, another on 58 and a final one on 59, the latter receiving a whopping 143 because they were out on their own yet, in reality, they actually only scored 3 more points than the children on 56 who ended up with 130.
To sum up, in the upper echelons of scores each mark extra scored is worth a considerable amount extra after standardisation.
But that is only my take on it. Possibly some statistician will come along and tell me I am talking rubbish.
I also agree that Concerto's daughter is exceptional.
What would be really interesting, however, would be the raw scores which yielded these marks.
I have seen quite a few raw scores for the Sutton exams and the affect of standardising does distort the whole final score issue. For instance last year I had one candidate on 44/45 in VR who scored 141 standardised and one who had 43//45 which standardised to 133. One point could be down to little age disparity but basically the first child received a whopping 8 more, simply for getting 1 question more right.
Extrapolating from this, if the upper scores for KE trailed off at, let's say, 56/80 in the maths, but there were a fair few grouped round this score, they might end up with a standardised 130. Then you might get a child on 57, another on 58 and a final one on 59, the latter receiving a whopping 143 because they were out on their own yet, in reality, they actually only scored 3 more points than the children on 56 who ended up with 130.
To sum up, in the upper echelons of scores each mark extra scored is worth a considerable amount extra after standardisation.
But that is only my take on it. Possibly some statistician will come along and tell me I am talking rubbish.
I don't think we're disagreeing, although we might be approaching it from different directions. My argument is that at such a high level, the odd mark extra/dropped here and there due to day to day differences in performance will have such an effect that you can't read very much into an individual score, regardless of what population it is standardised against. If you had results from a dozen or two such children in the two exams it might be possible to make some definite statements, but for that you'd need a test population well into the thousands. There aren't enough children taking the Warwick test to provide reliable results.
Mike
Mike
-
Charlotte67
- Posts: 893
- Joined: Thu Oct 11, 2007 8:59 am
- Location: Cloud 9
Um. When my daughter sat the NFER test (for entry in '08 ) I was assured by both NFER and the LA that the scores were standardised within the cohort only. No external data was used.
Where have you heard differently Ken?
Ta.
Charlotte
btw - I have always argued this point:
Where have you heard differently Ken?
Ta.
Charlotte
btw - I have always argued this point:
Mike wrote:There aren't enough children taking the Warwick test to provide reliable results.
