Quote:
Originally Posted by Doomfishy
@quasievilgenius - Science is indeed about statistical probability; there will be instances when 100% of the data aligns (yes, everyone who jumps will fall back to Earth), especially when considering basic physics and chemistry. It is relatively easy to control for confounding variables when you're working with inanimate compounds.
But there are many other cases when experiments show a causal effect because of the statistical improbability that the data is the result of chance.
Does smoking cause lung cancer?
Well, yes, medical science has definitely supported that conclusion. But it's definitely not the case that 100% of smokers develop cancer - it's just statistically improbable that researchers would have gotten the data that they did by chance.
When dealing with areas of science that inherently include confounds that are difficult or impossible to control for, considering statistical probability becomes a necessity.
|
Alright, that's a compelling argument, though I would like to say:
My reason for responding as I did is that your original statement syntactically appeared to suggest that science is derived purely from statistical probability, which is not really the case, and is an over-simplified view of science. I never said that statistical probability is not a scientific tool, merely that the whole of science is not based on statistical probability, it's based on arriving at an explanation for a phenomenon through observation. Statistical probability exists within the data analysis aspect of science, not as much on the data collection/experimentation aspect, as you have to have data to derive statistical probability, which is a distinction I should have made initially, as I fear now that we have been arguing opposite ends of the same side of an argument.