Support thread for parents of chilldren sitting the transfer test in 2011

Kola

RikM wrote: »

It doesn't work out that way...
Tests scored out of 64 mean that your combined raw score is out of 128.
90/128 = 0.703125 (call it 70%)

But that doesn't relate in any simple way to the standardized score.
Nominally that's out of 145 and 84/145 = 0.57931... (call it 58%).

But part of the standardization process resets the average mark on the AQE to 100... That's after they've adjusted the marks for age. So you would expect a child who scored average for their age group of entrants to have an AQE mark of 100 (or 69%, after standardization).

One of the reasons some of us are interested in the raw scores, is to see if there is any obvious logic to how the marks were arrived at.
So far, there's not enough information.

I cannot tell you how the standardisation process works. All I can do is quote raw score and standardised score in the hope this may provide some information for others.

Kola

This is an email from AQE 10 February 2012. Might be of some use to others.


We do not know where the vast majority of children's score's lay. With 100 as the middle mark there will always be a tighter number of children around this score and the further your move away from this number, the number of children decreases. (For example 30% of children scored between 99-109, 30% scored between 110-123 and 40% scored under 99). This is the nature of standardisation.

Unfortunately quintiles do not exist this year, for several reasons. While they were a useful guide for parents, however they were misused by some areas of the media and others, and turned into grades etc which was always inaccurate and caused more confusion.
Several post grammar schools had already published criteria stating that quintiles would be part of their admissions process and for that reason, for this year only, a guide has been provided for their information to help them with that.
In terms of understanding your score there are several factors to keep in mind;
1. The lowest scores accepted by the grammar schools you are interested in for the previous two years. This can act as a guide, though obviously the boundaries can vary slightly.
2. The middle mark is 100. A child with a score over 100 should have no trouble getting a grammar school place, keeping in mind point 1. A child with the score in 90's will also have a good chance of getting a place in a grammar school, though it does depend on the school.
3. The highest score acheived this year was 123, as you already are aware.

The top score acheiveable would not have been much more than 123, these children scored very highly. With standardisation, the bar is set at the middle mark and simply keeps going until the highest candidate's score is reached, in this case 123.

Third_Time_Mum

RikM wrote: »

It doesn't work out that way...
Tests scored out of 64 mean that your combined raw score is out of 128.
90/128 = 0.703125 (call it 70%)

But that doesn't relate in any simple way to the standardized score.
Nominally that's out of 145 and 84/145 = 0.57931... (call it 58%).

But part of the standardization process resets the average mark on the AQE to 100... That's after they've adjusted the marks for age. So you would expect a child who scored average for their age group of entrants to have an AQE mark of 100 (or 69%, after standardization).

One of the reasons some of us are interested in the raw scores, is to see if there is any obvious logic to how the marks were arrived at.
So far, there's not enough information.

I think the most interesting information relating to age standardisation we've seen so far is the raw scores compared to the AQE scores of Rik's and Becca's Mum's children last year. Rik's had 92% raw score and 118 standardised score with a February birthday. Becca's Mum's had 96% raw score and 117 standardised score with a birthday only 3 months earlier (in November). This is certainly a much bigger allowance for age than would have been given under the old transfer system. It would appear to confirm my suspicion that AQE standardises within the individual months that children are born, ie, they set the average mark of 100 for the children born in that particular month and then work up and down according to their raw scores. Thus, the standard for November must have been a good bit higher than the standard for February. That was my understanding of the reply AQE sent to Mumsicle back in October (quoted below from her post on 10 February) although I might have totally the wrong end of the stick!

"In our process we compute the age of every child to the nearest month using the child's precise age on the day of the test. This allows us to assign each child to a unique age category. A child aged 127 months goes into one category while a child age 128 months goes into the next category, and so on. No child is left out of the process, and children can change category as they progress through the tests. Our algorithm ensures that for any mark, the percentage of children scoring above that mark is precisely the same irrespective of the category to which the child has been assigned. In this way, every child is "competing" only with children of the same age (to the nearest month)."

What do other people make of this statement, particularly the sentence beginning "Our algorithm ensures ..."???

minnimo

Just thought I would post the results we had for AQE. My son sat the exams in Nov/Dec 2010, his raw scores were 62 53 and 56 and his AQE result was 117. His birthday is July so he was one of the older children sitting the exams. I have no idea re how standardisation is then applied, but I have another child due to sit the AQE exams this year who is a May birthday so will be at the other end of the age spectrum this time.

Pmum

Third Time Mum,

Did we not discover that Rik and Beccasmum's kids did the tests in different years so we can't compare the raw scores and AQE marks?

Pmum

Actually,

minnimo and RikM's kids did the test the same year from what I can see. They both scored 92% in their best 2 tests.

The one with the birthday in February got an AQE score of 118.
The one with the July birthday got an AQE score of 117.

7 months and only one AQE mark difference.

Third_Time_Mum

Pmum wrote: »

Third Time Mum,

Did we not discover that Rik and Beccasmum's kids did the tests in different years so we can't compare the raw scores and AQE marks?

Sorry to mislead anyone; when Beccasmum compared her child's score to Rik's I thought she was implying that they had sat the tests in the same year. You're right, you can't compare scores from year to year.

aqeworrier

Thanks for your words of encouragement Third Time Mum. I'm hoping scores will improve as time goes by.

I know its impossible to but I've been trying to make sense of the standardised score and percentage and using scores from this year - Kola's score of 70% with an AQE score of 84 and Confused Mummy's score of 86% with 102 AQE score I've made the following table which may give some guidance re score. I know it's very basic and subject to much criticism but its only meant to be a general guide - I note that they seem to go up by about 1-2 points per percentage? obviously there is age to be considered too - It will be interesting once other people get their scores to see how close they are to this.

Percentage AQE score
70% 84
71% 86
72% 87
73% 88
74% 89
75% 90
76% 91
77% 92
78% 93
79% 95
80% 96
81% 97
82% 98
83% 99
84% 100
85% 101
86% 102
and so on...

This would suggest that the average score was around 84% - pretty high in my opinion...

Pmum

aqeworrier,

i wouldn't get too hung up on trying to equate the scores your child is getting in practice tests to that table. It really does sound like the scores acheived in the real thing are quite a bit higher than the same kids are getting in the practice papers.

It would probably be a better indicator to ask their teacher how they think your child is doing compared to other children who have sat the tests in the past few years. Please don't panic if your child isn't scoring in the 90%s in the practice papers as, in my experience, very few children will be and yet more than 50% of the kids that sit the test will get the Grammar school of their choice.

Certainly I know that the average score in practice tests my childrens classes would not have been anywhere close to 84% and yet I am sure the average AQE score in the same class was significantly greater than 100.

flymetothemoon_2

I still think after all I've read on here that a child probably needs to be scoring in and around 70-75% to get an AQE score of around 100. Rik M's post 440 sounds about right to me. I also recall a teacher in my kids primary school telling me that the school reckoned kids needed to score around 66% or more to get into a grammar and that was based on several years of 11+ results. So, I would say to anyone who is worrying that their kids need a really high score of 85% to stop worrying because levels of achievement cannot have changed that much over the past few years and so it is still likely that around 66% + will hold good for most grammars.

My DS scored 72% in the 2010 tests and got an AQE score of 96
and if, as Rik M suggets we assume that the 96 is out of 145 then her score of 96 is actually equivalent to 66% and she did go to a grammar and could have chosen from others with her score of 96 so please don't think your child needs a really high score becasue they don't.