Age Standardisation

[Leighmum2019]

My DC scores
M- 48
E-48
SC - 369.56
Born month-- Sep

from the https://docs.google.com/spreadsheets/d/ ... edit#gid=0" onclick="window.open(this.href);return false;
I saw one score
M-47
E-48
SC -- 376.82
Born Month July

still cannot get my head around age standardisation

A child 10 months younger than yours got the same raw score. Both are impressive but child 2 has a slight adjustment to their standardised score to mitigate the fact that they are considerably younger.

[Hedgehogmum]

My DS
E 41/60
M 38/60
Dob : oct /07
Standardised score : 333.91

Point 29 on spreadsheet
E 42/60
M 38/60
November baby
Standardised score : 337.98

More than 4 points for 1 month and 1 mark in English.

Does this sound correct?

[diligent]

I do agree age standardisation should be considered in this kind of exam for young children (primary school age). Generally, academic ability (not knowledge) is different for children between ages 10 to 11. Older children do have some age advantage over younger ones. CSSE has tried to eliminate the advantage by applying age adjustment this year, but it seems to have gone too far. Now the age advantage has become a disadvantage. Also this year’s scores cannot be comparable with previous years’ scores, the Guidance Sheet may not be accurate.

Here is the math:
Two children: C1 and C2
C1: DOB 01/09/2007 (Older Child)
C2: DOB 31/08/2008 (Younger Child, 364 days younger)
Both have the same English and Math raw scores: English 40/60, Math 40/60
Both have the same standardised total score: 334.145

Based on CSSE’s method: https://www.csse.org.uk/images/simplefi ... -10-18.pdf" onclick="window.open(this.href);return false;

But after applying age adjustment (age adj.)
The older child’s (C1) total score (with age adj.): 334.145 – final score
The younger child’s (C2) total score (with age adj.): 344.806 – final score
Therefore, the younger child has a final score 10.661 higher than older child.

If the older child wants to beat the younger child’s final score, he need at least 5 more raw points in English to make the total final score 346.921. He/she would need to do 8.3% better to get an additional 5 points.

The mathematical model used by CSSE is:

Final score = (standardised Math score + ”days younger” x math age adjustment) + (standardised English score + ”days younger” x English age adjustment)
Where “days younger” is calculated as D-O-B minus 01/09/2007

This mathematical model is using the linear model, which assumes that a child’s academic ability increases daily with the same speed, but, in reality, the speed is not a constant, as the older you are, the less you improve per day. There is no mathematical model to 100% match the reality, but if you use the square root (SQRT) model below, it should be more accurate.

Final score = (standardised Math score + SQRT(” days younger” x math age adjustment)) + (standardised English score + SQRT(”days younger” x English age adjustment))

Based on this model:
Both have the same standardised total score: 334.145
After applying age adjustment (age adj.)
The older child’s (C1) total score (with age adj.): 334.145 – final score
The younger child’s (C2) total score (with age adj.): 339.736 – final score,
Therefore, the younger child only has 5.591 points higher than the older child.

If the older child wants to beat the younger child’s final score, he only needs 3 more raw points in English to make the total final score 341.81. He/she would need to do 5% better to get an additional 3 points. This is more reasonable and possibly makes the system fairer to both older and younger children.

[Hedgehogmum]

I completely agree with age standardisation but I wonder on the accuracy and measurement adjustments.
Also 4 points for one month and 1 mark difference seems unbelieveable.

[pinkrabbit38]

My DS
E 41/60
M 38/60
Dob : Oct 07
Standardised score : 333.91

Point 29 on spreadsheet
E 42/60
M 38/60
November baby
Standardised score : 337.98

More than 4 points for 1 month and 1 mark in English.

Does this sound correct?

That’s my daughter and she is right at the end of November so in effect 2 months difference

[DSDSDD]

My DS
E 41/60
M 38/60
Dob : oct 07
Standardised score : 333.91

Point 29 on spreadsheet
E 42/60
M 38/60
November baby
Standardised score : 337.98

More than 4 points for 1 month and 1 mark in English.

Does this sound correct?

Yes. 2.555 points for the 1 mark and another 1.519 points with a birthday of late Nov

Exact DOBs removed by moderator, to avoid identification Apologies

[Hedgehogmum]

Ok thanks everyone. Fingers crossed pink rabbit.

[Wontsleeptonight]

So at the KEGS post 11+ visit today they said (quoting from their slide which I took a photo of)
"Warning: the number of entries for the 11+ has gone up; the average score of candidates sitting at KEGS has gone up; and age standardisation will on average increase scores. We cannot predict actual cut offs but we believe that the cut offs for KEGS are likely to rise this year."

I am still waiting on the CSSE response which is disappointing given that they close for half term a d it's getting close to the submission deadline.
I am impressed with KEGS for identifying this shift and for their honesty.

Here is the slide:
https://drive.google.com/file/d/1g-85Qr ... p=drivesdk" onclick="window.open(this.href);return false;

[hessian77]

So, what is the practicality around all these? Let's summarise the evidence so far (and probably final one):

- there is a chance some thresholds may increase vs last 4 years, but this isn't sure and it isn't clear by how much
- some children may indeed be impacted e.g. the ones around the thresholds but not sure if in any case

Nothing mathematical or specific in all these, just an open door for speculations as we have seen plenty so far.

What is the practical approach of a child who scored 334 and is hoping for school X which had the last admission score around 333 in the last 4 years? Or for one who scored 331 and hopes for same?

Evidently, to keep applying for it they both have reasonable realistic chances to enter.

The only certain way of losing a game is by not playing it. Don't fall in the trap of not applying when you still have a chance just for giving some people who were below an additional chance.

[hessian77]

Even if you fail by 1 point - e.g. the threshold raises to 335 in my previous example of a pupil scoring 334 and being above last 4 years ones, the child's result remains still exceptional and he or she deserves congratulations. There is not a clear line between winners and losers as some may think. It isn't the end of the world, but make sure you have played all your cards.

On the other side, everyone who is in this situation will keep a fair '50-50%' chance to entry back of the mind and will consider a Plan B.

But my point is this chance will be 0 if you don't play it at all.