Over the last few weeks we have published a series of posts analysing the 2011 GCSE DfE data. We believe it shows that academies perform no better than comparable state schools, and on some indicators perform worse. We have refuted the DfE rebuttal. Gove advisor SamFreedman (@samfr) has engaged with us on Twitter but has declined to discuss it further. Two broadsheet education correspondents have offered to either have the rival views independently assessed or publish an email debate. @samfr has declined to take part and no other DfE spokesperson has volunteered.
It is possible that DfE spokespeople and the supporters of academies are simply very shy and like to avoid publicity. But it is hard to avoid the more likely conclusion, that they are unable to argue with the clear conclusion of the data.
Time to Fact Check
Independent fact checkers Full Fact have kindly offered to give their assessment of the viewpoints. So here I will set out the facts to be checked and invite the supporters of academies to put any rival views. Our claim:
The 2011 DfE data on GCSE results provides no evidence, despite DfE claims, that academies perform better than comparable non-academies. Indeed on some criteria they perform worse. This is despite the extra funding that these initial sponsor-led academies received.
This is the statement we would like to be checked. We are referring solely to the 2011 data, as this is the only data that has been released in sufficient detail to enable comparisons of comparable academies and non-academies. The DfE, in response, have claimed that the 2011 data paints a different picture and that academies grew twice as fast as non-academies between 2010 and 2011.
The key points of our case were put in this Observer article
, in the subsequent post
showing that the DfE claim that academies grew faster did not stand up (and that post links to most of the detailed analysis, also available here
), when they were compared to similar schools. The DfE set out its arguments in a response
to the Observer article published last week. We in turn published a post to refute that argumen
t. There was also an interesting bit of analysis of the 2011 data
by Leonard James which came to the same conclusion.
The key explanation of the different interpretation is contained in the House of Commons briefing
on the performance of academies, specifically the graph on p9. This shows that schools with low 2010 results showed strong growth (over 8% on average for those starting from under 35%) and those with high 2010 results (over 65%) showed, on average, no growth at all. Therefore any sample would show growth above the average if it had a high proportion of schools with lower figures for 2010, as most academies did.
The key is to ensure you compare similar schools. When we compared schools
whose GCSE results were under 35% in 2010, we found both academies and non academies showed an average 8% growth in results from 2010 to 2011. This is great news, it shows that overall schools that were getting low results are improving fast. But it makes no singificant difference whether they are academies or non-academies.
That is our evidence and I will tweet @fullfact asking them to check our claim. I will also tweet @samfr inviting him to put any contrary evidence and I also invite any supporters of academies to post their comments below.