A month long hiatus was never my plan but over that time I've had exams and assignments to concentrate on, unfortunately. So with those commitments over my head I couldn't find time to do adequate research for any posts. Yes, I do research for each post...all two of them. I couldn't find time for a lot of things in fact.
It seems I simply started the blog a bit too early. Now, though, I have a lot of free time ahead of me so I should be able to get a post out at least once a week. With pictures and links and other improvements. How exciting, eh?
Anyway, to get back into the swings of things I'll post a short little entry on an important aspect of the scientiftic method – the fact that correlation does not imply causation.
Simply put, this is a phrase used in science and statistics to emphasize that just because there's a correlation between two variables, this is not proof that one is the cause of the other. An idea that's simple enough to grasp but you'd be surprised how often people can be fooled and overreact when shown a correlation and told it's the cause. An asshole called Dr. Wakefield used a certain correlation to imply that a Autism was caused by the MMR vaccine. I'll go into that a little at the end.
But before I do, I'll just outline a couple of examples where if you were to use the logical fallacy - “correlation does imply causation” - how you'd look like an idiot.
During either the Korean War or the Vietnam War, I can remember which, in one base there had been an insane amount of crashes involving soldiers on motorbikes. The army conducted a study to see if they could find a reason, a commonality between all these soldiers and why they were crashing their motorcycles.
The results of the studies... Tattoos. The single thing all people involved in these accidents had in common was tattoos. The idea that it was the tattoos causing the crashes was ridiculous. It could be argued that people who get tattoos are more likely to ride motorcycles and take chances. And people that are more likely to ride motorcycles and take risks are more likely to crash motorcycles. But to say a tattoo is the cause. You'd be a fucking retard.
In a widely-studied example, numerous epidemiological studies showed that women who were taking combined hormone replacement therapy (HRT) also had a lower-than-average incidence of coronary heart disease (CHD), leading doctors to propose that HRT was protective against CHD. But controlled trials showed that HRT caused a small and significant increase in risk of CHD. Re-analysis of the data showed that women undertaking HRT were more likely to be from higher socio-economic groups, with better than average diet and exercise regimes. The two were coincident effects of a common cause, rather than cause and effect as had been supposed.
Thanks to wikipedia for that example.
Now, to talk about Wakefield who conducted a study on the links between the MMR jab and Autism. When his reports were released in 1998 implying that the jab was in fact a cause of autism people went bat shit retarded and refused to give their children the MMR vaccine. A truly stupid decision that borders on child abuse and I say that with minimum hyperbole.
The findings are now widely believed to have no credible scientific basis at all but we still have people who refuse to give children the MMR vaccine. You can read more about the whole controversy here.
Thing is, if you approach cunt Wakefield's findings using the scientific method (read common sense), it falls apart pretty quickly.
The use of the jab while Wakefield was doing his research was widespread – the odds that the autistic children that Wakefield was using for his study had got the shot were huge. This doesn't mean the shot was the cause. There would have been a load of other commonalities between these children that meant just as little.
People seem to refuse to think a lot of the time.
No comments:
Post a Comment