Page 1 of 2
another question about standardisation - sorry
Posted: Sat Jan 10, 2009 9:09 pm
by Bougalou
If a child got a higher raw score in the second test, but a lower standardised score than the first, am I right in thinking this shows that most others in the cohort did better in that test?
B x
Posted: Sat Jan 10, 2009 9:59 pm
by Guest55
Yes it must have been 'easier'.
Posted: Sat Jan 10, 2009 10:31 pm
by KenR
That's why you need to find the Standardised Score for each of the test components as that will give you an accurate assessment of the percentile that your child achieved in that particular test component.
If you take a look at the Standardised score section in this web site
http://www.elevenplusexams.co.uk/standa ... scores.php
and in particular the NFER paper
http://www.nfer.ac.uk/research-areas/as ... -ranks.cfm
That shows the percentile score for each standardised score.
Hope this helps
Posted: Sat Jan 10, 2009 11:42 pm
by Bougalou
thank you
Posted: Sun Jan 11, 2009 12:41 am
by Bougalou
One last question, - pretty sure I didn't need to do any calculations as I have all raw scores and standardised scores.
I just need someone else to confirm for me that 5 standardised marks equates to fewer raw marks, i.e. if a candidate was out by 5 standardised marks it actually means that they missed the cut off score by less than five raw marks. - hope this makes sense and thanks to those who have answered so far.
B x
Posted: Sun Jan 11, 2009 12:41 am
by Bougalou
One last question, - pretty sure I didn't need to do any calculations as I have all raw scores and standardised scores.
I just need someone else to confirm for me that 5 standardised marks equates to fewer raw marks, i.e. if a candidate was out by 5 standardised marks it actually means that they missed the cut off score by less than five raw marks. - hope this makes sense and thanks to those who have answered so far.
B x
Posted: Sun Jan 11, 2009 12:28 pm
by KenR
Hi Boulalou
Not necessarily, every Age Standardisation is different - it varies with with exam type, cohort and the difficulty of the test.
If you want to see some real world examples of Age Standardisation take a look a the following spreadsheets published on the QCA web site. These relate to years 3, 4 & 5 testing and Key Stage 2 tests
http://www.qca.org.uk/qca_8772.aspx
http://www.qca.org.uk/qca_14158.aspx
http://www.qca.org.uk/libraryAssets/med ... d_2005.pdf
http://www.qca.org.uk/qca_14151.aspx
If you search around the site you will find other examples.
Just a point of clarification, in many of the tables at the low and high end you will see that no score is given and that *** is used.
You should read the corresponding quote given
Very low and very high standardised scores are printed in the tables as ***. This means that they would be below 70 or above 140, but cannot be calculated with the necessary degree of statistical reliability. If an exact score is needed, for example to calculate an average for the class, 69 or 141 should be used for these children.
This explains for example why there are often a lot of scores of 140 or 141 in 11+ tests.
Hope this helps
Posted: Sun Jan 11, 2009 3:57 pm
by Bougalou
Ken, they have been really useful thank you, I need to adjust my speech as a result
B x
Posted: Sun Jan 11, 2009 6:27 pm
by Guest55
Bougalou,
I would not focus too much on the number away from a 'quaification' score. I'm sure you will spend 95% of the time on the academic evidence that proves the scores were a 'blip'.
Good luck
Posted: Sun Jan 11, 2009 11:04 pm
by Bougalou
Yep, I'm doing just that Guest 55,- have some evidence in from invigilator's report which I wanted to link to smallness of gap between achievement and cut off - was trying to make gap as small as poss!!
Bouga