Friday, October 23, 2009

Urban myth hits the internet. Media coverage ensues.

What can you not prove with a little mathishness? A good question.
About a month ago there was a little coverage about a pharmacy put up its online calculator that tells you the number of indirect sexual partners you've had. What is an indirect sexual partner it's someone you've not had sex with but one of your sexual partners has. Which basically is that old urban myth twaddle about you're not just sleeping with someone, you're sleeping with everyone they've ever slept with, only with a place to modify the numbers.
People take some stuff really seriously. Like these guys: The Independent. The Christian Institute. This dude. And at the risk of channeling Ben Goldacre form a moment, the common factor is that they lazily breathlessly copied the press release without even so much as a challenge to the basic assumption or the execution.
So let's whinge for a moment.
First execution. You enter your current age, number of sexual partners and your gender. Then you select the age of each of the sexual partners. The calculator uses the average number of partners for someone that age and then applies the Kevin Bacon principle by calculating back six 'generations,' again by average number.
What it doesn't seem to care about is overlapping networks. If you go out that far, a lot of the people are doubled up. Someone that you slept with has slept with someone who has slept with someone who has slept with someone you also slept with.
In other words if Kevin Bacon is in a movie with Goldie Hawn and one with Kurt Russel he's been in movies with two people not with four.  Add Kate Hudson to the mix. A movie with Kate and one with Goldie gives you 3 (including Kurt in the indirect mix) people, not six, Goldie plus 2 indirect and Kate plus 2 indirect.
So there's a built in bias to exaggerate the numbers. It's all about overlapping sets.
But the assumption is sold as STI (STD, VD or whatever they're calling it now) awareness. And this is where it really fails.
One because it doesn't tell you anything about your actual risk but instead tries to bamboozle you with big numbers. (Actual Scaremongering Headline: Average Brit adult has indirectly slept with 2.8m people! 2.8 million! Jesus Titty-Fucking Christ, that's more pussy than Wilt Chamerblain.) But what does that tell you about risk?
Why? Because safe sex with a hundred people puts you at very little risk, while unprotected sex with 3 hookers in a Bangkok back-alley puts you at a hell of one.
So what is the real assumption?
Basically it's devoted to the idea that there's an inherent taint to sex that is passed from person to person. Otherwise there's no reason to consider 'indirect sex' with someone who hasn't passed a disease on.
Imagine someone telling you that you're not just shaking hands with a person, you're also shaking hands with everyone they've ever shaken hands with. You'd think they were nuts, wouldn't you? Try it again with 'hug.' Sounds every bit as ridiculous, wouldn't it?
What what we have is pseudo-scientific moralizing...
Load of crap.


Anonymous said...
This comment has been removed by a blog administrator.
Anonymous said...
This comment has been removed by a blog administrator.
Sicilian said...
This comment has been removed by a blog administrator.