Tuesday, 22 May 2007
The abuse of statistics
I noticed this little piece that paints doctors as being dangerous because they are 'more lethal' than gun owners. This is argued because in Sweden one is 7500 times more likely to be killed by a doctor than by a gun owner. What marvellous logic.
It is also true statistically that one is more likely to be killed by a family member than by a serial killer. By the same logic this would mean that our family members are more dangerous than serial killers. Something does not quite add up here.
Obviously statistics are being abused here. It is very rare for someone to be gunned down in Sweden, and it is also rare to be blown up while rolling a large bomb down a hill;
"(2006, Vietnam) In a similar event, a Rolling Stone isn't all that gathers no moss. Three men scavenging for scrap metal found an unexploded 500-pound bomb perched on a hill, and decided to retrieve it with help from Sir Isaac Newton. As they rolled the bomb down the hillside according to the laws of gravity, the bomb detonated, leaving a four-meter crater and sending the three entrepreneurs to a face-to-face meeting with their Maker."
Obviously one's chances of being killed by a doctor are much higher than being exploded by rolling a large unexploded bomb down a hill, however this does not make rolling large bombs down hills safe and doctors dangerous.
I don't even have to explain this any further, it's pretty damn obvious that millions of people are seeing their doctors as we speak and statistically this means that some of these people will be being harmed by their doctors as we speak. However this does not make doctors dangerous, in fact the work of medics the world over will be doing a rather large amount of net good. Conversely the infrequency of people being killed by unexploded bombs does not make these bombs safe; it's just that people aren't often stupid enough to try rolling them down hills!