Darby
Epochal Historian
They get a fair hearing. No one stops them from telling their tale.All I am saying is that people deserve a fair hearing, and you don't seem to approve of that.
The scenario is this: They have a right to tell their story. I have a right to listen but no responsibility to listen. If they want to be believed they have a responsibility to convince me, assuming that they aren't simply barking at the moon or crasting about the tired and boring Titorism, "I don't care if you believe me." (Which begs the question why post or tell the story if you don't care to be believed...but that's another topic.)
Now to put it more in terms of the scientific method:
They are telling a fantastic story. The null hypothesis is their story is not true. Considering that most all the time all such stories are found to be untrue we might set our confidence level at 99% therefore alpha = .01. This means that one out of every hundred times that we "run the numbers" on our experimental design for such fantastic stories we will make a type 1 statistical error . We will in error reject the null hypothesis and accept the story as true (false positive).
Frankly, in these cases alpha should be .001. Doing so means that we will inappropriately reject the null hypothesis 1:1000 times. But there is a trade-off. By setting alpha so low we increase beta - the probability that we will make a type 2 error and accept the null hypothesis when the story is actually true (false negative). Considering the historically low probability of such stories being true I'm willing to accept a type 2 error over a type 1 error. That's a choice that every experimentalist has to make with every experiment run.
Last edited by a moderator: