Tuesday, 21 April 2009

Happy talk - behind the UK's 'dismal record' for children

A few years ago, a report appeared from UNICEF which claimed that Britain was the worst possible place for children to live: the whole thing was lapped up gleefully by everyone from the Daily Mail (as a stick to beat Labour) to Polly Toynbee (in her crusade for more cash for early years). Now a 'new' report for the Child Poverty Action Group - surprise, surprise - says pretty much the same thing and is being given top billing on Today and in the Mail. Today's Mail said that the CPAG research 'echoed' the UNICEF report. Well, it would do, wouldn't it, since the principal author of both reports is Professor Jonathan Bradshaw of the Social Policy Research Unit at York University who decides which criteria to use in drawing up the league tables.

Now, having read some of his work, I think Professor Bradshaw is pretty good at his job. He can write reasonably well aside from occasional lapses into jargon - which can't always be said for academics - and has developed a perfectly reasonable theory about how to measure relative childhood happiness and quality of life across countries. But it is just that: a theory. Yet it is being treated as irrefutable fact despite his league tables (as with the UNICEF report) giving equal weight to different things that few would regard as of equal importance or reliability. The CPAG report doesn't tell you anything about Professor Bradshaw's weighting of different aspects; in order to understand why the UK does so badly, one has to understand his weighting. To do that you need to fork out $34 for an article he wrote with Dominic Richardson of the OECD (in an independent capacity) in a journal called Child Indicators Research, published earlier this month.

The research splits 43 indicators into 19 components across 7 domains, giving each domain equal weight. There, you find, for example, that the reason the UK does poorly on child health is because of its relatively low childhood immunisation rates. These are given equal weight to infant mortality and to something called 'health behaviour' based on questions to children about whether they brush their teeth, are too fat or eat apples. The UK is apparently doing fairly well on these latter scores, but has been downgraded by its low immunisation rates. I wonder which newspaper and which morning current affairs programme has most to answer for on that score?

Then, there's something called "subjective wellbeing" - which is ranked as importantly as health or education. This has three components - each one three times as important to a country's score as whether kids can read properly. These are "personal wellbeing - the percentage of children reporting high life satisfaction" - whatever that means; wellbeing at school - whether children "feel pressure at school" (bad, apparently) or "like school a lot" (good); and 'self-defined health" - whether children think they are healthy. Each of these three subjective components is ranked more importantly than whether a country's babies die in infancy. Anyway, needless to say, British kids score below average on this lot. But it is comforting to learn that in league table-free Finland, regarded as the best education system by many, children feel just as much pressure and are just as likely not to like school as their British counterparts. The authors tell us this is because "educational attainment may be a well-becoming indicator rather than a well-being indicator." I did warn you Prof Bradshaw was guilty of occasional lapses into jargon.

Next up is relationships. There are just two components here, each worth considerably more to a country's ranking than good literacy or low infant mortality. And as Britain is slightly above average here, we probably shouldn't complain. But it is again subjective and is based on the percentage of children who "find it easy to talk to" their parents and who find their classmates "kind and helpful." In case you were wondering, France is not a good place for this sort of thing. I trust President Sarkozy has a taskforce on the case already.

Then we turn to material wellbeing, which is based on relative poverty indices, measures of deprivation and children in workless households. This seems, like health, to be a reasonably objective indicator. Children in the UK are, apparently, among the most likely to be living in workless households, but the authors tell us this does not mean they are lacking in consumer durables (colour TVs, computers or cars) or under severe economic strain. The next 'domain' is risk and safety. The three components here are 'violence and violent behaviour' (fighting or experiencing bullying), child deaths and risky behaviour (early intercourse, smoking, drugs, drunkenness). Surprisingly, perhaps, the UK does a bit better on this list, being brought down by youthful drunkenness, but having a relatively low number of child deaths.

Then we turn to education. The UK gained slightly above average PISA scores (among the countries in this report) in literacy, numeracy and science (though the combination of these is only worth the same as each of the subjective questions above). The other two components are educational participation of 15-19 year-olds and in pre-school and NEET rates (those not in work, education or training). Since there is an obvious correlation between post-16 participation and NEETs, the authors have chosen to give half of the education score to this aspect of education. There is no mention of university participation, where the UK does well. And the relatively high pre-school participation rates, where the UK is also doing much better than many - and which most researchers would regard as the most crucial element - are apparently worth just a third as much as 16-19 participation/NEETs. That is the authors' choice but this is not an accurate representation of any education system. There is a final domain for housing issues, which is a second poverty grouping.

In their academic article, the authors themselves show how easy it is to manipulate the data - and, to be fair, are happy to offer it to anyone else wanting to do so. They pick just seven indicators - child immunisation, 'high life satisfaction', talking to dads, lack of educational possessions, recent bullying, maths scores and houses with housing problems and the UK finds itself up in 18th position instead of 24th, ahead now of France and Italy.

I have no quibble with researchers reporting these indicators and highlighting where Britain ranks according to each one of them. I also think it is important that young people's voices are heard - as is increasingly the case in schools. But I worry about arbitrary weightings which give far more weight to subjective - and perhaps culturally sensitive - questions than to matters of life and death or pretty basic educational outcomes. The CPAG should publish all this information on its website, but if the media weren't so keen on talking our country down, they might actually explain the arbitary nature of these rankings - and own up where our low rankings reflected their own efforts rather than the policies of the government.

This post has been picked up by John Rentoul.

1 comment:

Sam said...

Great rebuttal article Conor. Also reminds me of the article Robert Hill wrote a couple of weeks back demolishing the "independent schools get more As than comprehensives" claim. There needs to be more of this more forensic examination of the facts, rather than taking them all at face value.