Origyptian Wrote:
-------------------------------------------------------
> Simply because when you're only looking for the
> highest and lowest value within a data set, other
> statistics that further characterize that data set
> are irrelevant. Only when you try to make
> conclusions about those values do statistics
> become relevant, such as "are they from the same
> population?", or "are they outliers?", etc.

Quite. They’re relevant to such questions as, are the data good or bad? Are they a suitable basis for further inferences?

Don’t know about you, but it’s the further inferences interest me. Why else bother?

Jon is trying to bypass the tests, but still make the inferences. Why aren’t you objecting?

Where are these “higher standards of proof” I keep hearing about?

M.