(June 11, 2011 at 3:50 am)tackattack Wrote: I was using that one partiicular study (I'm sure there are more, it was just a simple google high hitter) to support my assertion that statistiacally there is not zero evidence that ESP breaks probablity.
How can you possibly use one study to demonstrate something statistically?
(June 11, 2011 at 3:50 am)tackattack Wrote: If that view were valid, studies like this wouldn't be possible.
Of course they would, that's the point I was making in my last paragraph about confirmation bias.
Imagine some normally distributed random variable, X, with mean <X> = 0. Now suppose that this underlying distribution is unknown to me, and I make the hypothesis that <X> > 0.
Now suppose that I make 100 trials, each consisting of 100 measurements of X, which correspong to one measurement of <X>. Given that X has mean 0, on average I will find that <X> > 0 50 times and <X> < 0 50 times.
Now suppose I say "aha! 50 successful trials!" and submit the results of those 50 trials for publication in the The International Journal of X Studies. Could that be considered evidence in favour of the hypothesis <X> > 0 ?
Galileo was a man of science oppressed by the irrational and superstitious. Today, he is used by the irrational and superstitious who claim they are being oppressed by science - Mark Crislip
