Does water make you smarter? QTWTAIN

'How's that? Would a glass of water help? Scale of 1-10.'

Is it silly season again? Oh yes, it's always silly season when it comes to the latest social science confection about education. And this one's an old one, with grey whiskers dangling around its ankles: the claim that drinking water raises exam grades. Or, to be more specific, that taking water into the exam hall may improve grades by 'as much as five percent.'
Can you hear that rapping sound? That's me, drilling my head into the table like Woody Woodpecker. Statements like this keep my holy mission lamp burning- to expel as much wooly-minded bollocks from the already heavy burdens of teachers as possible. But oh, the bollocks falleth passing hard, and often.

Let's take a slightly closer look at this, in the manner of a post-mortem crime scene investigator. Which is appropriate, because this theory is ripe, mature as a good Roquefort. Some of you may have heard of Brain Gym, one of the most knuckle-headed moronisms ever dropped in our classroom laps. It believed, among other things, that touching your pressure points 'activated the brain'; that doing special exercises made your thinking 'clearer'; and that the brain had to be 'hydrated' to operate at peak efficiency. I'd also like to point out that its proponents believed that the water was best absorbed through the roof of the mouth as it was 'a more direct path to the brain'; presumably they had fathomed an as-yet undiscovered route between the palate and the medulla oblongata that mere empirical science has overlooked. The stalwarts who advocate this latest research aren't making such a claim, luckily. Just thought I'd mention it.

Brain Gym was adopted and paid for by thousands of gullible, well-meaning schools, and in some areas (such as my teacher training) directly sponsored by an equally gullible DfE, which apparently believed that 'anything's worth a punt', in that way that it bloody well isn't.

So, to the present claims. The Universities of East London and Westminster researchers found that, of 447 psychology students, one quarter of them took water into the exam hall. So, around 112. Of those 112, examination grades were on average 4.8% better across the cohort. Conclusion: drinking water may have been responsible for this.

That 'may' is very important. Because the researchers, not being idiots, haven't simply claimed that A (water) caused B (5% increase in results); they suggest that it may. Big difference, and quite right too. Also, we have to be careful, in a way that the media coverage of this wasn't- every news outlet I saw today reported the big 'Water increases grades by 5%' headline, with weasel apostrophes to bury their honour. Press on science stories leans- always- to the big claim, so we can excuse the researchers that charge. But here are the problems anyway:

1. How do we know that the students did 4.8% better than they would have normally?

We don't know what they would have obtained without the water, for the simple reason that we don't have the ability to divine what would have happened in alternate dimensional realities of our own Earth. Maybe they would have obtained exactly the same grades. Maybe not. Who knows? That's the point: who knows?

Of course this kind of data is usually predicated on other data; coursework, previous test scores etc. Usually what happens is that candidates prior attainment data is compared with other canidates eventual attainment, and probability graphs are produced, with estimates of where the candidate should probably be as they proceed through their courses.

'Glug...glug....getting smarter...'

The huge, huge, leviathan problem with this is that they simply aren't reliable. Just because most people who obtain, say 65 % on their course work later go on to get (on average) a 2:2 or whatever, doesn't mean that any individual candidate will. I've seen- often- kids jump from Es and Fs at GCSE to Bs and Cs at A-level, because of a change in attitude, work ethic, family situation etc. And vice versa. This deterministic model of human behaviour might satisfy the current bean-counter's passion for cramming the human experience into statistical models, but it is about as smooth a fit as a yoghurt pot on the horn of a rhinoceros.

So: we don't know what would have happened without the water to those students AND we can't predict the future. Those are pretty big problems. Yet some people in the social science faculties wave them off as irrelevances,as if they were minor details. They aren't. They are enormous. They are cosmic. They are the problem.

2. How on earth do you know the water had any effect at all? The problem that people like Feynmann used to delight in pointing out to social scienctists, is that human behaviour is so complex, so resistant to simple reductivism, that anyone trying to point to one cause leading to one effect in behaviour, is going to have a very hard time indeed. If I want to see why water is getting hotter, I can investigate the factors; I can remove factors one at a time; I can control. I can blind, I can double blind, whatever I want to do to make the experiment as clean as possible. Turns out it was the bunsen burner. Who knew?

It's so very, very hard to do this with humans. A given factor (say, giving a specific stimulus, like saying 'You look fabulous') can have an enoprmous variety of effects (a phone number; a slapped jaw; an invitation to audition for La Cage aux Folles). Because there are countless influences on human behaviour, and that even before I get to the nebullous, wondrous possibility of a free human will (Heaven FORBID). This is often referred to as a high causal density; in other words, there are too many possible influencing factors on human behaviour to be able to pick out any causal relationship with anything other than the force of conjecture.

So maybe the water did help. Maybe bringing water in reveals that the candidates are nervous, and need some kind of totem (has anyone done any research on the effect of gonks?) that reassures them; one that says, 'Yes, my brain shall not wither and dessicate in the space of the next two hours. I can relax.' Maybe the water drinkers were hungover; maybe THAT helps. Who knows? Not I. And not this research.

What is the possible mechanism of this wonderful claim? Is it that you can't think straight when your brains are dry? Is it that water is magic? How quickly does it take for imbibed water to reach the brain and meaningfully have any kind of interaction? Surely the exam would be over by then? How much water needs to be ingested?

Honestly, this is so daft that I can barely support my lower body. Sure, I can probably have a punt and say that being refreshed and fed is a good place to be for a student in an exam, but what more can be extracted from this research other than those experientially obvious claims? Nothing. Nothing meaningful; conjecture, and a pitiful conjecture at that. A hundred or so students were studied. A hundred? I could find you a hundred students who do better after a packet of Monster Munch. Give me thousands; give me wave upon wave of hydrated ubermensche and I will start to accept what you claim. But a hundred? F*ck. That.

I wouldn't mind, but this is exactly the shade of guano that rains down on us in the education profession, regular as Sunday. Money is spent, time is blown, and directives of best practise are issued. It is an enormous, self-copulating industry of sloppy thinking and waste. And we have to deal with it, every rainbow and unicorn flight of fancy that someone with a clip board and a hard-on dreams up.

We do not have time for this shit. All it does is keep me in good blog material. I could live without it.