Friday, 20 May 2011

Value Added Scores.

I am intrigued at this report on the BBC Wales website dealing with school's performance based on Value Added Scores:
Schools are scored based on whether each pupil's GCSE results are better or worse than what was predicted of them at the age of eleven. The measure singles out schools which enable pupils to exceed expectations, whether, for example, it is an E grade student who achieves a C, or a B grade student who achieves an A*.

This Value Added league table seems to confuse and contradict rather than clarify issues of performance.

Take for example schools in Blaenau Gwent and the Vale of Glamorgan. Blaenau Gwent have the highest spend per pupil whilst the Vale of Glamorgan has the lowest. GCSE results in the Vale of Glamorgan are the highest with 73.4% achieving A* to C with Blaenau Gwent on 48.7%. The Value Added scores shows that Blaenau Gwent with six secondary schools have two schools with positive scores, none with negative scores and four with a neutral scores. In the Vale of Glamorgan only one of eight schools had a positive score and again no schools having a negative score but seven having neutral scores. Interestingly enough Blaenau Gwent are in 3rd place in the Value Added Score table with the Vale of Glamorgan in 10th place.

The situation in Wrecsam is similarly clouded with the following rankings:

25. Ysgol Bryn Alyn +

35. Ysgol Morgan Llwyd +

42. Ysgol Grango 0

45. Ysgol Rhiwabon +

59. Ysgol Rhosnesni 0

84. St Joseph's 0

119. Darland 0

175. Clywedog -

211. Maelor, Penley -

Compare this to last year's GCSE results where 61% of Wrecsam pupils achieved 5 A* to C but came in at 9th place one ahead of Vale of Glamorgan and six behind Blaenau Gwent! Three schools in Wrecsam saw massive improvements in their thresholds for A* to C:

Darland 77% +12%

Rhiwabon 65% +22%

Bryn Alyn 56% +9%

There seems to be a correlation between value added scores and GCSE results for Ysgol Morgan Llwyd at the top and Ysgol Clywedog at the bottom. There seems to be no correlation between Value Added scores in respect of Darland School in particular.

I would very much appreciate views on these various anomalies and contradictions and what exactly do value added scores tell us about our schools.

2 comments:

Anonymous said...

You cannot expect a year on year increase in GCSE results to correlate with value added since it shows whether a school has performed better or worse than expected with regards a particular cohort.

The value added also takes into account performance in english/welsh + maths, whereas I doubt the percentages you are quoting have done the same.

As for the vale versus Blaenau gwent, yes a traditional league table on GCSE a to c would show the vale in front. What the value added shows is that pupils in schools in Blaenau gwent are on the whole exceeding expectations, but this may well be D or E grade students getting a C.

The data actually rewards schools who are having a positive impact on pupil outcomes but get overlooked when you look simply at raw GCSE exam data.

It makes no sense at all to compare a school in the richest part of the vale with a school in the poorest part of BG on exam results only as the kids entering secondary schools in the vale would always be more likely to achieve the higher grades.

Plaid Gwersyllt said...

Thanks for your contribution, but your comment just goes to show the complexities and the dangers of just considering one data set. I wonder exactly what Leighton Andrews' new league table will actually measure?