Of the 249 schools listed in the DfE data as 'sponsor-led' academies, almost a third are run by chains. As Fiona Millar pointed out in the Guardian last summer
, these are one of the fastest growing aspects of our new education landscape. The Oasis chain has grown from £3 million to £70 million between 2006 and 2010 and Ark from £3 million to £117.5 million. The E-act chain grew from £15.5 million to £60 million in just one year, from 2009 to 2010. (Since Fiona published this information, the chains have chosen to make their accounts private and so future figures will not be available.)
The chains appear to be strongly favoured by government. Toby Young, for instance, has commented
that future free schools will only be able to get through government procedures if they are run by chains. E-act alone plans to run 126 schools
by 2015. So how are the chains performing?
The Telegraph reported last week
that "academies inflate results with easy qualifications". They found that while the results of non academies fell from 59% to 53% when "equivalents were removed", those of academies fell twice as much, from 50% to 38%, a drop of 12 percentage points.
Our analysis shows that the figures for the academy chains fell still further, from 52% to 36%, a drop of a full 16 percentage points:
The two most highly regarded chains fell most: ARK schools fell from 64% to 3%, a drop of 21 percentage points and Harris schools fell from 68% to 53%, a drop of 15 percentage points.
Once equivalent qualifications are removed, every chain achieves significantly lower results than the average non-academy, with four of the seven chains having an average below 35% achieving 5 GCSEs including English and Maths. E-act, despite its ambitious expansion plans, achieved an average across its schools of just 29%. Ormiston managed an average of just 22% [Figures/chart corrected, 9/3/12]
The chain schools do tend to have more disadvantaged intakes and so a better measure of their success is value added. Here there is a stark difference in the performance of the different chains. There is a good reason that the success of the ARK is often quoted, as their record is strong. However it is not matched by the other chains. If we look at value added for progress in English, where 1000 is the national average, only ARK schools are above this national average:
On average chain academies score 998.5, below the national average, for value added in English. For value added in Maths they are at the national average. A similar picture emerges if we look at % of students making expected progress in English and Maths. In both categories the ARK and Harris chains are above the national average but the other five chains are below it. And overall the chains are again below the national average.
The only categories in which all the chains perform strongly are those, like Best 8 value added, where the use of GCSE equivalents boost the results. These have been much criticised by the government and it is interesting to see that the groups making most use of these equivalents are the chains that are being strongly encouraged by the same government.
The DfE data makes clear that government policies are driven by ideology and not by evidence. If it were evidence-based, there would be major doubts about conversion to academy status and even greater doubts about encouraging the chains.
Note: These figures include only those schools which have reached GCSEs. Around 15% of the chain academies had no GCSE results for 2011 and are not included in this analysis.