A Room With 100 Seats
14 August 2003
I'm now 83% of the way through writing this 'clown. You see, that's the kind of thing I get paid to come up with, because people like to see things put into numbers. My guess is this is because people in business, and life in general, like things to be simple, clear-cut, black and/or white. Numbers are simple, so people are comforted by them. Numbers are good.
So, when someone asks me how far through my 'clown I am, I don't count from the first word of the final version, I express my progress as an estimated percentage of the overall process of 'clown-writing. Now, for me, that includes thinking of some sort of idea, deciding that it's not very good so trying to think of a better one, not succeeding and facing stretching some tenuous spark to a bag of words, several days of procrastination, followed by writing it at work while pretending to be doing some vital documentation or something. So according to my measure, I'm virtually finished. Unfortunately, however, up till now I've only done one hundred and eighty four words, and that includes title and sub-title. I don't have much to show for my 83%. Numbers are highly ergonomic - easily and comfortably manipulated.
Words as well as numbers suffer this exploitation. Take for example the old chestnut of "virtually fat free". Since "virtual" means "not real", then surely the meaning is "not really fat free". You're not going to buy their unit trust if their "market-beating returns" are "not-really guaranteed" are you? But the marketers are saying something that at the same time is technically true, but sounds better than it really is.
Numbers do this particularly well. 90% fat-free sounds better than 10% fat, but means exactly the same thing. Several well-meaning government agencies have decided to try and cut down on this sort of thing, but it's not going to be possible. You can't stop people thinking up ways of making their product sound better than someone else's by blinding the public with statistics. The theory is that if five out of six dentists or seven out of eight midwives use it on their own families, then you should to. And no doubt the company will have signed affidavits from six dentists and eight mid-wives to prove it, because that's all they'd need.
If anyone out there can explain to me how hair can be sixty percent shinier, please let me know. I would like to know what the SI unit for hair shininess is, and how it was measured. And it's not just "makes hair 60% shinier!", it's always "makes hair 60% shinier!*". Once you've found the small print and borrowed an electron microscope, you see that this claim is from clinical trials comparing peoples hair washed with this stuff against people who washed their hair with water taken from the sewage plant outflow of the North Dakota Municipal Sewerage Board (Looking after Number One - and Number Two).
You think I'm making this stuff up. Two real examples from respected news publications follow:
Among three-year-olds however the prevalence in males was running at 16%, twice that in females.
But is the prevalence in females 8% or 32%? An ambiguous sentence of the worst order. The whole point about dealing with numbers is that they make the world seem more certain, not less. Find this person and slap 'em for me.
Put another way, in a room of 100 people, only 13 will be left-handed. In the same room, nine will be left-handed and gay and three will be left-handed, gay and missing a limb. Spare a thought for them.
I had to read this again. Firstly, thank you dear article writer for condensing the idea of a percentage into a room of 100 people. That 13 are left handed I accept wholeheartedly, and imagine that it has been researched very thoroughly, although it is not clear if this is a national average or global, but that's beside the point. But nine left handed and gay? Assuming that there is no relationship between handedness and sexuality, then if you had a room of a ton of south-paws, 69 of them would be gay, which seems like a lot to me. Left-handers bat on the other side, not for the other side. And if you had a room with 100 gay left-handers, 33 of them would be missing a limb. Or possibly more accurately in that room, 33 and a third. That's one unlucky demographic.
Now it's one thing to go and quote numbers willy-nilly, but to do so and do it wrong is even worse. Simple numbers always conceal complexities and ambiguities which is bad enough, but numbers that are patently wrong mislead. But they're likely to get past (let me generalise here) a bunch of editors who haven't done any maths since sixteen and were probably shit at it then too.
All in all, numbers are powerful things that should be wielded responsibly. When someone throws a number at you by way of proof, it seems harder to argue back. But behind every number is a thousand arguments, assumptions and inaccuracies, so take every one with a heavy dose of salt. I should know. Over the past three years, 48% of my time has involved constructing statistics to back up arguments. Therefore I know.