This is the analysis on which today’s TES article
on primary performance is based.
The LSN analysis of secondary performance found that, once compared to similar maintained schools, sponsored academies improved their results no faster than maintained schools. Analysis of primary performance produces a different and startling result: When compared to similar schools (based on 2012 results), sponsored academies grew at a slower rate than maintained schools:
It is important to remember that primary academies have not been piloted in anyway. There is no prior evidence base for the effect of converting an “under-performing” primary school to an academy.
Instead a huge experiment is taking place across the country, as primaries seen as in difficulty are encouraged or forced to become sponsored academies. Almost 2,000 primary schools have now taken this step into the unknown.
My estimate below is that a primary school staying in the maintained sector improved 5% faster
– on average – than one becoming a sponsored academy.
There is a massive success story here, with which the DfE should be very pleased - and would be if it was committed to the success of all schools. In 2013 there were 316 primaries that grew their SAT results by at least 20% in 2013 (including only those with more than 25 pupils in Year 6): 94% were maintained schools, 5% were converter academies and just 1% were sponsored academies. The full list can be downloaded here
The DfE ArgumentThesis
: "Under-performing" primaries perform better if they become sponsored academiesEvidence
: SAT increase of sponsored academies is 3x that of all maintained primaries
These primaries perform much better if they stay as maintained schools.Evidence:
When compared with similar schools (ie, with similar 2012 SAT results), maintained schools outperform sponsored academiesPolitical implication:
Given a primary school at a particular level (and below the 80% benchmark) it is likely to perform better if it remains as a maintained school than if it becomes a sponsored academy.
Data: Schools grow more when they start from a lower base
I don’t dispute DfE’s claim. Sponsored academies grew by over 3% in their Year 6 SAT results last year, compared to 1% for all schools. (The data used throughout this post is the % of pupils achieving level 4 in reading, writing and maths.)
However there is a clear inverse relationship between previous year SAT results and growth in those results.
So if you take a set of schools (such as sponsored academies) primarily from the high growth bands they are going to have higher growth than schools overall – even if they are underperforming within their bands.
Maintained schools consistently outperform similar Sponsored Academies
The graph at the start of this article compared compared sponsored academies with maintained schools in the same 2012 bands, for those below 80%: Here are the underlying figures
(The under 20% band isn’t included because there are only 10 schools. The over 80% band isn’t included because there are only 10 sponsored academies and because the key issue is under-performing schools. The change in both groups are equal in the 80%-100% band at -4%.)
| ||Number||SAT growth 12-13|| |
|2012 Range||Sponsored academies|| Non academies|| Sponsored academies|| Non academies|
This seems to show fairly strongly that maintained schools do better, when compared to schools with similar 2012 results.
Maintained primaries grew 5% more
The first graph showed the average growth in each band. This can be used to calculate, for every primary school in the country, whether their change in results is above or below this average, and by how much.
Using this, overall maintained schools grow 0.1% more than the average and sponsored academies grow 4.9% less. So the cost of becoming a sponsored academy seems to be 5% in SAT growth. (This differential grows to 5.3% if only those schools previously under 80% are included.)Recommendation: Conversion of "under-performing" primaries to sponsored academies should be suspended while this effect is examined. It is possible that primary schools, generally much smaller and with less resources, than secondaries face greater challenges if they move away from the local authority.
This analysis refers only to "sponsored academies". Schools are generally converted under a sponsor if they are deemed to be in need of signficant improvement. The analysis does not included "converter academies", which are generally schools already seen as Good or Outstanding.
This analysis is based on three datasets from the Department for Education:
2013 KS2 performance tables: http://bit.ly/KOepvQ
. This includes KS2 results for each school in 2012 and 2013. However for most primaries that became academies in 2012-13, it does not include the 2012 results of the “predecessor school”. This makes the above analysis a little complex to carry out and is why, I presume, it has not been done previously by others.
Open academies: http://bit.ly/1fStloT
– this dataset enables the academy to be linked to the predecessor school
2012 KS2 performance tables: http://bit.ly/1hQLK6j
Only sponsored academies that had converted prior to 1st January 2013 were included in this analysis.