Monday 3 June 2013

The Sutton Trust - don't make me laugh

The Sutton Trust, an educational charity, has come up with a corking demonstration of the woeful blindness that blights UK educational policy.

It reports today that top state schools "are significantly more socially selective than the average state school nationally and other schools in their own localities", with a much lower proportion of kids on free school meals.  Affluent middle class people are crowding round the best schools.  The Trust thinks that "schools, particularly in urban areas, should use a system of ballots - where a proportion of places is allocated randomly - or banding across the range of abilities to achieve a genuinely balanced intake".

But the Trust is mistaking correlation for causation.  It hasn't asked two fundamental questions, the first being "what is a good school?", the second "how does a school become good?"

The Trust defines good schools by results, referring to them as "the highest performing comprehensive schools".  Good results = good schools.  Let's assume that's right (although I personally think "good" is better defined in terms of value added).  Now how did those schools become good?  Obviously a lot of different things go into the mix - quality of leadership, quality of teaching, quality of facilities; but surely the one which trumps them all is quality of intake (I know the plural of anecdote does not equal data, but my own kids have done very well in a comprehensive in which some of the teaching was terrible and the buildings so depressing and squalid that they would have been a disgrace in 1970s Albania).  If a school has an intake in which bright industrious kids well supported at home are over-represented, it will tend to do better than one in which they're under-represented.

So in which section of society are bright industrious kids well supported at home most likely to come from?  Why, the middle class of course.

Keep with me here - I'm not suggesting that all working class kids are thick and have feckless parents, or that all middle class kids are all paragons; I'm just suggesting that bright industrious people tend to marry people like themselves, and either become or remain middle class.  Not surprisingly, given a triple whammy of genetics, culture and affluence, they tend to raise children who take on their parents' characteristics.

It may not be an especially appealing thought, but we should not be surprised to find that the bright and industrious are over-represented in the middle class compared to the working class.  This can only be not true if a) intelligence and industry are not heritable qualities, or b) we live in society that does not reward brains and effort.  They are; we do.

So actually the Sutton Trust has got it risibly the wrong way round.  There is a link between the middle class and the best performing state schools, but it doesn't work the way the Trust imagines.

The Trust thinks the middle class cluster round the best schools.  Actually the best schools are best because the middle class cluster round them.

We will never get educational policy right in Britain while we suffer from this delusion - and Michael Gove suffers from it as well - that all wrongs can be righted if only we can make all schools "good" schools.  The Sutton Trusts goals are actually more limited, in the sense that it is only trying to equalise access to "good" schools.  But it too is chasing a shadow.  If you equalise access the only result will be that "good" schools will start to get slightly worse results and "bad" schools slightly better.

The Trust says, comically, that "lower income students do better when there is a mix of students of all backgrounds in a school".  I'm sure they do.  The corollary must also be true - that higher income students do worse; and moreover as you reduce the number of higher income students in a school, the lower income students, previously doing better because of them, start to do worse as well.  The Sutton Trust are strangely silent about this.

No amount of tinkering will fix the basic problem, which is that a lot of kids aren't very bright, and in any event come from households with a range of dreadful social problems, not the least of which is that the parents didn't get anything in particular out of the education system themselves.  The kids are fly enough to know that they aren't going to be lawyers or doctors, but not quite clever enough to be realise that education is not just about money.

What might just make a difference, because of the very large numbers of children involved, would be the abolition of private education.  A tsunami of Tarquins and Tamsins would descend reluctantly on the Gasworks Comp.  Then you would see some equalisation.

But which government is going to risk that one?

P.S.  A  very good letter in the Graun the following day from a fellow sceptic wondered whether affluent middle class people enjoy better health because they are treated by "top doctors", or whether leafy suburbs enjoy lower crime rates because they are watched over by "top police officers".

Very pithily nailed.  Bah.  Humbug.