More school statistics...

... which most people would have guessed anyway if they thought about it for a few moments, and which can be used to support any particular point of view/prejudice:

From The Telegraph:

Choosing a nursery could be more important than secondary school when it comes to ultimate academic achievement, a study has found. The contribution of a school to variation in pupil attainment is in fact “relatively small”, according to researchers at University College London's (UCL) Institute of Education. 

I wonder when another university will do the same analysis of whether universities make much difference either, for a given standard of student...

This means that sending children to what they regard as a “good” secondary school - meaning one that fares well in GCSE results and official progress measures - only adds a small amount more value than if they went to a “bad” secondary school.

The report, funded by the Nuffield Foundation, found that once you control for the characteristics of pupils within each school, the difference in attainment linked to school is below ten per cent of the total variance between children.

Professor Alex Bryson, UCL’s Chair of quantitative social science a co-author of the report, added: “Schools do still matter, but they don’t matter as much as people think. What perhaps is slightly surprising is that an awful lot of time and effort goes into trying to improve schools – but over the 13 year period we looked at, the variability accounted for by schools has remained fairly constant.”

Which would happen if all schools were to improve by the same amount, so this does not mean that schools haven't improved.

But:

Researchers found that school competition, measured by the number of schools within 5.5 kilometres of each school, accounts for up to seven per cent of the variation in pupils’ attainment.

That could be 'competition driving up standards' (hooray), or just as possible, slightly cleverer kisd getting into the perceived better schools and vice versa, leading to a widening gap between schools perceived as better or worse.

Meanwhile the pupil-teacher ratio - an indication of how well resourced the school is - only accounts for around one per cent of the variation. A report last year by researchers at King’s College London found that genetics are a bigger determinant of academic success than sending your child to a grammar school.

We could have guess the last bit, it's genetics and parents pushing their kids to do their homework etc.
-------------------------------------------------------------------------
On the subject of competition between schools and/or competition for better pupils between schools, from the BBC:

Most families do not choose to send their children to their nearest school, shows the biggest ever study of state secondary school choices in England. More than 60% opt for a school that is further away - usually because it is higher achieving.

"Contrary to a widely-held belief, only a minority of parents choose their local school as their first option," say researchers.

Why would parents not put down a perceived better school as their first choice? Be daft not to; and what are the chances that the perceived best school within reasonable travel distance is not the nearest one? Fairly high, I would have thought. Bear in mind my point above that there is likely to be a wider perceived gap between two nearby schools than between two isolated schools which all children in an area attend, simply because there are is no realistic alternative.

It also debunks the idea that richer families are more engaged with choices... Despite any assumptions about the "sharp elbows" of middle-class families, there was no significant difference in behaviour between wealthier and more disadvantaged parents.

Again, that doesn't really prove anything.

If a school is perceived as good, then higher income families will choose to move near it and will drive out lower income families from the area. This leads to the self-enforcing gap between the areas. Higher income families, having paid over the odds to access the 'free' education they've paid for two or three times over (income tax and higher house prices) will of course send their children to the nearest school - because that one is perceived as best (or else they wouldn't have moved there etc).

The reverse applies to low income families where parents 'just want the best for their children'. They can't afford to live close the perceived best schools, but they can still put them as first choice on their applications.

And so on. All quite fascinating in a Freakonomics kind of way.