Why everyone is talking about London schools
Another week, another report on London schools. This time, it's by CentreForum, a Liberal Democrat-aligned think tank. This is the third investigation published in the past fortnight into the happy mystery of why the capital's schools are now so good.
Work from CentreForum is worth noting for two good reasons. First, whether or not one agrees with its conclusions, the think tank is particularly strong when it comes to analysing schools. It has an unusual strength in statistics.
Second, the body is closely linked with David Laws, the schools minister, and Tim Leunig, the Department for Education's chief analyst. The think tank, above all, has nurtured the government's secondary school league table reforms.
That new league table is rather complex. Normal tables ask, "How well did the pupils at each school do?" Instead, the new tables will ask, "How well did secondary schools do, given how well prepared their pupils were at the age of 11?"
This new "progress" or "value-added" measure is a strange hybrid: as well as measuring school performance, it encourages schools to take traditional subjects. A school cannot do well, even if pupils get straight As, with too many vocational subjects.
This new report has set out its estimates for how schools would do under this league table, were it applied to the latest results. It is slightly unfair to do this: after all, schools were not aiming to do well on this measure.
Even so, it is a useful exercise. The report strongly supports the notion, which I have discussed before, that something exceptional has happened in London secondary schools in addition to its primaries.
The CentreForum measures imply that - looking only at traditional subjects - a child at a school in Westminster, Hackney and Islington would beat a similarly able 11 year-old in Hull, in south Yorkshire, or Knowsley, in the North West, by more than a grade in every subject when they finally took GCSEs.
Some of that is down to that curriculum choice. Schools in London have tended to choose a tougher, more traditional curriculum than other parts of the country. They happen to be doing a mix of subjects that is rewarded by the measure.
So the gap ought to close as schools outside the capital respond to the new league tables. Some schools are simply putting children into qualifications that do not show up on this measure. But some of the difference is not about subject preferences.
For example, if you look solely at English and maths, back in 2012 pupils in London schools beat children who were similar at 11 by a third of a grade in both subjects.
As subject choices start to shift, that simple metric might be the measure to watch to see whether schools in different parts of the country are actually improving, or whether they are just changing subject choices.
The London advantage is deep, too. As the IFS report last week stated, London's raw results are boosted by its primaries, too. Those London children who took GCSEs in 2012 were already ahead at 11.
In fact, they were nine percentiles higher than children from similar social and economic circumstances in Yorkshire when they started secondary school.
So a child in Yorkshire whose background would suggest that they would expect to finish halfway along the league tables could, had they lived in London, have been expected to finish 41% of the way down the table.
So what does that all mean for England's schools? Many of the report's recommendations are all about replicating the London Challenge, a scheme to encourage school collaboration and support teachers in the capital.
You can expect a lot more "how can we be like London?" if the results keep restating the capital's advantage.