SW Herts consortium schools: admission statistics

[WP]

The Watford Grammars allocate 60% of their specialist places to an inner catchment area and the other 40% to an outer catchment area.

Next year it is 65%/35% (Source: http://www.hertsdirect.org/docs/pdf/m/mon2012dir)

Well spotted. I've updated the text.

[AlisonR]

Yes, great spot. I hadn't noticed that either.

That's 65% of the 45 academic places to the Inner area instead of 60%. I make that 29 instead of 27 for the inner leaving 16 instead of 18 for the outer area. So two more academic places for the inner area and two less for the outer area.

(update) Hmm... just looked at the September 2012 entry admission rules for both Watford Grammar School for Girls and Boys and it seems the 60:40 split is still there. There is a 65:35 split mentioned between Community Places and Specialist places, with Specialist places being the music and academic places.

Alison

[WP]

(update) Hmm... just looked at the September 2012 entry admission rules for both Watford Grammar School for Girls and Boys and it seems the 60:40 split is still there. There is a 65:35 split mentioned between Community Places and Specialist places, with Specialist places being the music and academic places.

Yes, the full determined arrangements on their websites (and the Herts list) and the summary in the Moving On book are different. I understand that the determined arrangements are legally binding, so we must treat the description in the Moving On as erroneous.

[maddad]

Something just occurred to me about the standardised score: are the scores standardised against the general population or against the set of pupils sitting the tests?

The reason I ask is taking Parmiter's as an example:

Candidates sit tests in mathematics and verbal reasoning, which are then standardized and added. (NFER has a partial account of standardization.) Here are the lowest standardized scores allocated places under the academic criterion at each point:

Code: Select all

   2005   |   2006    |   2007    |   2008    |   2009    |   2010    |   2011    |
Mar   Sep | Mar   Sep | Mar   Sep | Mar   Sep | Mar   Sep | Mar   Sep | Mar   Sep |
----------+-----------+-----------+-----------+-----------+-----------+-----------+----------------------
...
  -   225 |  -    238 |  -    237 | 243   241 | 246   244 | 250   245 | 247   241 | Parmiter's
...

Assuming an average of 200 and a standard deviation of 30, we can translate standardized scores into the percentage of the children from an ideal population who would have attained at least that score:

Code: Select all

        0     1     2     3     4     5     6     7     8     9
----+------------------------------------------------------------
...
240 |  9.1   8.6   8.1   7.6   7.1   6.7   6.3   5.9   5.5   5.1
...

For example, a score of 235 places one in the top 12.2% of the population. 282 is the highest possible standardized score.

Parmiter's minimum score of 247 would imply the top 5.9% of children got offered a place in March 2011. But is that the top 5.9% of all Year 6 children, or the top 5.9% of children who sat the test? I expect that to matter, because children applying for a place on academic ability presumably (on average) come from the higher end of the academic spectrum.

I always use to think it meant the top 5.9% of all Year 6 children (and the text above says "the percentage of the children from an ideal population"), but I saw there are apparently 1101 pupils sitting the test for Parmiter's 2012 intake (http://www.parmiters.herts.sch.uk/filea ... t_LIST.pdf) of which (based on a quick count) it appears 746 are in the catchment area.

Now there are 44 academic places available to children in the catchment area, so the percentage of children taking the test who would expect to be successful is 44/746 = 5.9% (rounded to 1dp) - a coincidence???

I know it doesn't make any difference at the end of the day, but I'm interested in numbers and wondered if anyone else on here had an opinion (or knows!). I hope I've made sense...

[WP]

Something just occurred to me about the standardised score: are the scores standardised against the general population or against the set of pupils sitting the tests?

I think it's against some general population, but I don't have anything to quote that from.

Parmiter's minimum score of 247 would imply the top 5.9% of children got offered a place in March 2011. But is that the top 5.9% of all Year 6 children, or the top 5.9% of children who sat the test? I expect that to matter, because children applying for a place on academic ability presumably (on average) come from the higher end of the academic spectrum.

Remember that the same test is taken by all children applying for Queens' or Bushey Meads. (Queens' use it to get mixed ability classes, while Bushey Meads use it for their banded admissions.) Each child gets a single score, which is then used by the schools they've applied for.

[maddad]

Remember that the same test is taken by all children applying for Queens' or Bushey Meads. (Queens' use it to get mixed ability classes, while Bushey Meads use it for their banded admissions.) Each child gets a single score, which is then used by the schools they've applied for.

The question is whether the list on Parmiter's website (with 1101 children) is just the children that have applied to Parmiter's. I suspect so, since the other Herts consortium schools seem to have their own different lists, including Queens' (http://www.queens.herts.sch.uk/Docs/Sec ... s2011.xlsx). The list from Queens' has some overlap with the Parmiter's list, but less than 50% (and probably just represents the children who've put both schools down).

[WP]

The question is whether the list on Parmiter's website (with 1101 children) is just the children that have applied to Parmiter's.

It is just the applicants to Parmiter's, but that's not the question. People who put down both Parmiter's and Queens' do the test once, and get a single score, which must be independent of how many applied to a particular school.

[maddad]

OK, I see what you mean now. My theory would give them different scores depending on the school they were sitting for.

I think what I'm having trouble reconciling is that (in my very limited experience), it's typically only the "top table" (for example 7 children in my DS's class of 30) who take the test for Parmiters and it appears from the list that there are only places for 1 in 17 of those (based on 44/746). That seems significantly worse odds that the 247 normalised score would imply (1 in 17 of 'all children').

[nickyroberts]

HI
thank you for this forum, it has been so informative. I have got Parmiters cut off scores for 2012
academic = 247
music = 37
distance = 1319.341 metres

they said the academic was out of max of 300, I always thought it was 282 - has it changed this year?

thanks
Nicky

[maddad]

they said the academic was out of max of 300, I always thought it was 282 - has it changed this year?

I'd be surprised if their tests were granular enough to distinguish between 282 & 300!

Do they also have a cut-off score for the out-of-catchment academic test?