The minister, misleading claims and the unfortunate truth: non academies improve faster than similar academies
As the Education and Adoption Bill goes through the Lords, the government continues to justify it its claims for sponsored academies. For instance Lord Nash claimed, in a letter to peers on 21st October, that "primary sponsored academies are improving faster than all state-funded schools". At the committee stage of the Bill in the Commons on 7th July a similar claim was made and this debate (extracted from the full discussion) took place: Nick Gibb: "Those underperforming schools that have been converted to a sponsored academy have, over a four-year period, seen their grades rise by, on average, 6.4 percentage points compared with 1% for local authority-maintained schools in the same period." Kevin Brennan: "Is he comparing that improvement with figures for schools in similar circumstances that have adopted other means of school improvement?" Nick Gibb: "My understanding is that those schools that have been sponsored academies for four years have improved their grades by about 6.4% compared with local authority schools over the same period." (ie, No) For the last four years, since the data was first released, this has been the standard claim of government ministers. The improvement in sponsored academies is compared to the improvement in all schools, despite the fact that sponsored academies are starting from a lower base and therefore are more likely to improve.
If this discussion were in health
Let us imagine a similar discussion in health: Minister: "Our new type of hospital results in sick patients improving their health at a faster rate than the overall population." Questioner: "Are you comparing these patients to patients in other hospitals, in a similar condition?" Minister: "No, we are comparing their improvement to the improvement in the overall population." Questioner: "But that is silly. The overall population is, on average, pretty well and not likely to improve their health further. Surely you should compare your patients to sick patients in other hospitals." Minister: "No, we prefer to compare them to the overall population. And the health of sick patients in our new hospitals improves more than that of the overall population." Questioner: "Is the idea that if you repeat it enough, people won't notice the flawed comparison?" Minister: "Patients who are sick and in our new type of hospital improve their health at a faster rate than the overall population." This would be absurd. And Nick Gibb and Lord Nash must both know this. The question is not whether an underperforming sponsored academy can improve faster than schools that are already achieving well for their pupils. The question is whether, as Kevin Brennan asked, a sponsored academy results in improvement "for schools in similar circumstances that have adopted other means of school improvement".
DfE accepted in the High Court that the valid comparison is with similar schools
The government knows that this is the only valid comparison. In the High Court in June 2014 the DfE accepted that those that start from a low point get the largest increase: “Mr Herrington, Director of the Academies Group within the Department for Education, in his second witness statement states (at para 15) that the Secretary of State does recognise that the improvement in attainment seen in schools is related to their starting point, and in general, those with the lowest initial attainment will see the largest improvements.” Source: Judgement, Warren & LB Barking v Sec of State, para 55, hearing: 25 June 2014 Accepting this point the DfE produced analysis comparing the performance of sponsored academies with maintained schools starting from a similar point. On the basis of 2011-12 data, they claimed not the five times or six times higher result but a "marginally higher" increase. That was at a time when the GCSE benchmark included "equivalents" such as Btecs.
Why do non academies perform so much better than similar sponsored academeis?
In more recent years the DfE cannot even make the "marginally higher" claim for sponsored academies. Whether you look at 2014-2015, 2013-2015 or 2012-2015, if sponsored academies are compared to similar non-academies then the non academies consistently perform better. Here is the comparison for 2012-15: Is it time to start asking what it is about maintained schools that make them so much more effective, and what we can learn from that for the underpeforming schools in the academy sector?