Lord Nash misleads peers on "coasting" schools
In advance of the second reading of the Education and Adoption Bill in the House of Lords, Lord Nash has sent a letter to peers, which includes this sentence on the measures in the Bill to tackle coasting schools: “Our focus on coasting schools is about identifying and helping those schools that may be achieving respectable results, but which are not ensuring pupils reach their full potential over time.” It is not clear whether Lord Nash realises it, but this explanation bears little or no relation to what the government is proposing as its definition of coasting schools. It would be an accurate description of the old Labour attempt to identify “coasting” schools. This defined a coasting school as one which achieved benchmark results above a figure like 60%, but where pupils made below average progress.
The Conservative definition is the opposite of what Lord Nash has described
A key part of the proposed Conservative definition is that the school has below average KS2 or GCSE results. It is not clear why this is a “coasting” school or how Lord Nash can describe it as “achieving respectable results”. The definition is clear – it is about schools not achieving “respectable” results. A grammar school where pupils achieve Bs or Cs instead of As will not be identified as coasting. A secondary school with an intake strong enough to get 80% through 5 GCSEs, but which only achieves 60% will not be identified as coasting. A secondary school that achieves valued added in the top 20% of the country but has below average GCSE results could be a coasting school, despite doing more than either of the previous two examples to help pupils "reach their full potential". There are at least 20 schools that meet that last criteria that have valued added in the top 20% in the country but, with below average GCSE results, would be classified as "coasting" on their 2014 results. These 20 also have below average "expected levels of progress" in both Maths and English, because this measure relates more closely to their starting point than to their progress.
"Expected levels of progress" is not a value added measure
The "reach their full potential” suggests use of a value added measure, that accurately reflects the progress of all children. However the proposal is instead to use the deeply flawed "expected levels of progress" measure for English and Maths. This states that the "expected levels of progress" is 3 levels from age 11 to age 16 (ie, level 4 at KS2 to a C at GCSE, or level 5 to an A). It sounds reasonable but the likelihood to achieve 3 levels is hugely dependent on the KS2 starting point. For pupils achieving a 5a at age 11, fully 99% of pupils achieve three levels of progress. But only 43% of those achieving a 3a, and 15% of those with a 3c, get three levels of progress. (I give secondary examples but exactly the same is true of the "expected levels of progress" in primary schools.) See "How to use data badly" for more on this. Secondary schools will be classified as “coasting” if their GCSE results are below average and if the “expected levels of progress” are below average, taken over three years.
Is Lord Nash deceiving deliberately or by accident?
So it doesn’t address schools achieving "respectable results" and doesn’t use a valid measure of whether pupils achieve their "full potential". Instead, for secondaries, it will overwhelmingly affect schools who have a pupil intake with below average results at age 11 – even in some cases where they achieve above average value added for these pupils. This raises an interesting question. Does Lord Nash realise that the government's coasting criteria bears no relation to the definition he has given to the Lords and he is deliberately deceiving his fellow peers? Or does he himself not understand what the government is doing? Note: This description may change when the measure switches to Progress8, although we cannot yet be certain. Although it is a value added measure, there are suggestions it will also be biased in favour of pupils with higher KS2 results. See this from Education Datalab