Re: Official Trek XI Website Has Teaser Trailer

"Jaxtraw" <jax@[EMAIL PROTECTED]
> writes:
>
> And it's worth remembering that if I show that eating more than 4
carrots
> per day causes gout in women over 65, and you show that eating more than
7
> carrots per day causes gout in West African men between 40 and 60, that
> second experiment is not a replication of the first one. They're
entirely
> different experiments. In the ideal world, if you've found a statistical
> correlation of that type, the next step should be to use it as a
prediction
> and repeat the experiment to see if you get a similar result.

> But that's hardly ever done with that kind of epidemiology, because
> doing that would show that the overwhelming majority of such results
> are just statistical noise, and nobody wants to discover that, do they?

I think a literature search (let me suggest just in the last 2 years
in the area of bioinformatics) will suggest that people are very
eager to discover exactly that.  (Umm, because they are trying to
look for true answers?)