how much does age affect the result?

[KenR]

I agree with Mattseuf and fm, there are differences but this is a proven approach to compensate for maturity.

I did obtain a copy of the age standardisation conversion table a few years ago, the following is an extract of a posting I made at the time:-

The conversion of raw marks to standardised scores for a particular Grammar School will vary with Age profile - the difference can often be quite significant.

For example, in the Birmingham KE VR Tests in 2004 , to achieve a Standardised Score of 118 a child aged 11.18 years on the exam day would have to achieve score 71/100, whereas a child who was only 10.22 years would only have to score 65/100. Quite a difference.

You should also note that the raw 'pass' scores are significantly lower than the previous NFER tests (and most of those used in other Grammar Schools around the country).

So Mattsurf's figure of 2-3% is a good approximation at the extremes of Age

[tranquillity]

I can understand that it's only fair that allowances are made for younger children when exam marks are standardized ( as the mother of 3 summer born children, I know they are often more immature than autumn born kids, especially during their primary years!), but I'm still quite shocked at just how much of a difference the age weighting makes

[mattsurf]

I would think that the statistical confidence of the standardisation tables must be pretty high - ability for a given age would be normally distributed and the examining boards must have loads of raw data.

I agree that the range may seem a little surprising, however, I have no reason to question its validity.

[chad]

but I'm still quite shocked at just how much of a difference the age weighting makes

The age weghting is applied by cohort. Therefore if the July born cohort achieved the same raw scores as the October born cohort then there would not be any difference in the standardisation.
It is not a figure that is fixed before the exam.
Unfortunately (as said before) maturity does help at this age and inevitably there is a 2-3% difference between the cohorts at the upper and lower end of the school year therefore standardisation is usually applied.

[tranquillity]

Thanks to all you kind and patient maths bods out there, I think I've finally understood, at least in part. Please can you let me know if I'm starting to get the hang of this.

My understanding so far is that a September born child is compared with other children also born in September: their raw scores are all standardized together. Similarly, scores of June born children are standardized along with the raw scores of other June born children.

So, in the end, even though there is a difference in the standardized scores of a September born child and a June born child, this makes no difference to whether or not they are given a place at a grammar school. What is being looked at by the schools is what a child's score is compared to other children the same age as them. (I assume that this is what is meant by the term 'cohort' - a new word to me.)

The result it that age-in-the-school-year makes no difference as to whether or not a child gets a place because the only thing that's important it's how well they've done in the tests compared to children the same age as them. The cleverest ones, whatever their age, are offered the places.

Have I passed???

[mattsurf]

The result it that age-in-the-school-year makes no difference as to whether or not a child gets a place because the only thing that's important it's how well they've done in the tests compared to children the same age as them. The cleverest ones, whatever their age, are offered the places.

Have I passed???

Full marks - go to the top of the class