Statistics, smacking and bad science.

smacks I spent a good deal of yesterday ranting under my breath about correlative and causative data.

Why? Because of this story in the Daily Mail about how children who are smacked grow up to be more successful. Of course, being the Daily Mail, the story was followed by hundreds of comments from people saying that of course this was the case, smacking children made them better people.

But there's a HUGE difference between correlation (two events that tend to happen together) and causation (two events, where one is directly caused by the latter). So it may be that children from homes where they were smacked ended up more successful, but this doesn't mean the smacking was the cause of this success.

A great illustration of this idea is used in Freakonomics. American politicians looked at children's educational success and what was happening in their early years. They found that children from homes where there were lots of books tended to perform well at school. So they spent billions of dollars on a pre-school programme that gave books to poorer children, presuming this would improve educational performance. And it didn't make a bit of difference. 

The reason? Because the fact of books being in the home and the higher educational achievement of those kids was correlative not causative – meaning one didn't cause the other. In fact, closer analysis of the data showed that children who came from homes with lots of books out-performed other kids even when their parents didn't read those books to their kids.

So rather than books making kids smarter, it was more likely that the parents who bought the books were both smarter and wealthier, and were also the sorts of parents who were more likely to stress the importance of education to their children. (Incidentally, this very same study showed smacking or not smacking made zero difference to outcomes).

Despite all the claims you might read on the children's educational
activity websites about how their classes promote this skill or that educational outcome, I've yet to see a single one of these companies provide ANY kind of
statistical proof that these benefits are real. And certainly not attributable to their services.

Actually, I suspect it's like the books example – parents who take their children to educational classes are pre-disposed to think education is important and those children would tend to perform better on average, anyway. Certainly my experience of taking Flea to one of those classes when we first moved to Lytham was that it was full of middle-class kids who had absolutely zero need of a class to boost their social and communication skills. Pfft.

What this means for parenting – and yes, I do have a point – is that you can relax. Most of the things that will have a major influence on the outcome of your child's life have already been set. Your age when your child was born, your socio-economic status, your level of education – all of those things have a far more significant impact on your child's life than whether you took them to the right pre-school or whether they're learning phonics at the age of two. Really, what's the point? Isn't it better that they should just have the opportunity to, say, play? Have fun? Be kids?

Leave a Comment

Your email address will not be published. Required fields are marked *