You are viewing a single comment's thread from:
RE: How to live healthy? Part 4: How can I get to the truth?
first: great job as allways!
statistics in general is not about showing relationships. Its about reducing noise and taking out the annecdote from a statement.
one can easily have two not correlated parameters and will get a correlation according to p-value. The smaler the sample the easier it is. So over one half of findings is likely to be false and only noise. But since there are big trials with thousands of individuals, there is signal in science. But probably they realy should abandon the p-values from smal studys.
Hey lauch,
thanks for your reply.
I think this is worst of the worse!
How can someone know whether two changing parameters are correlated? How can they exclude that other parameters are responsible and the presumed correlation is just, as you said, "noise"?
The p-value suggests a scientific fundament which doesn't exist.
That's why I wrote it's important that also the readers should use their brains.
Let's see where this ends.
Best
Chapper