Remark & standardisation, is it different for VR and NVR

[tim101032]

We had our sons Slough consortium paper remarked and there was a mistake in the NVR paper which bumped him up a mark.

What intrigues me is that his VR raw score was much higher than his NVR raw score (by 7.5 %) yet the standardised score (SAS) was higher for NVR?

Would anyone know whether the method of age standardisation (SAS) is different for the Slough consortium for VR compared to NVR?

[aliportico]

I would imagine that it's not so much an effect of age standardisation, but that lots of kids did well on the VR, and fewer did so well on the NVR. So your son did relatively better on NVR than he did on the VR.

[tim101032]

I wont use the real figures but for example:-

If the scoring for my son looked like as follows:-

Verbal Reasoning Raw score 69 out of 80 (= 86%) - Standardised score 106

Non Verbal Reasoning 63 out of 80 (79%) - Standardised score 106

Overall mean percentage 82%

Would everyone expect that the age standardisation should give the same percentage reduction for Verbal reasoning as it does for Non Verbal Reasoning?

Is there another statistical rationalisation going on and are Verbal reasoning and Non Verbal reason raw scores treated differently?

Perhaps everyone taking the Slough consortium did fantastically well in Verbal reasoning which meant a reduction in scores was required?

I would like to understand this nuance especially as a mistake had been made with the marking.

When you have a remark is the standardisation process redone if not is it worth me asking for this to be repeated?

I assumed that the pass mark for the Slough Consortium was approximately 111/141 = 79% but this does not seem to be correct as it would appear age standardisation (in 2013) can reduce a child's overall scoring by up to 7% or more in NVR or VR.

Has anyone else out there had a remark done and found similar differences in scores?

[scary mum]

Here is an explanation of standardised scores
http://www.elevenplusexams.co.uk/advice ... xplanation" onclick="window.open(this.href);return false;

The two different tests would be handled separately. In simple terms your DC's score are only compared with children all born in the same month - so a bell shaped curve would exist for each month, standardised do that x% achieve a qualifying score. If, for example, all June born children scored very high marks on NVR, the pass mark for NVR would be high. If the same group of children scored poorly on VR the pass mark would be low. It would be interesting to know about standardisation if a remark changes the score - I've never heard of a remark changing a score before. I only know the basics of standardisation don't know. I assume that one more mark for one child would make no difference at all overall as the numbers are so large.

[tim101032]

In the case when a remark has found a mistake is it then worth having the age standardisation and scoring standardisation redone/checked.

If I request for this to be redone is it conducted by the school, the consortium or the examiner?