Rush? In defense of John

[Andrew Gilham]

>I understand that Andy Gilham
>earns his crust as a statistician.  Perhaps he can apprise us of whether
>John's sample size was large enough a sample size to extrapolate from?

Duh - what was the question again?  I did run some simple tests on the last
hawkpoll, and as I recall, while Space Ritual is unarguably better than Yuri
Gagarin, it's anyone's guess which is the best Weird tape.  (Nevertheless, I
enjoyed the hawkpoll immensely!)

Assuming BOC-L is a representative sample of anything but itself is a
decidedly dodgy assumption, in my view!  (For a start, because the hawkfans on
the list are fans of *current* HW - fans who didn't jump ship after Lemmy or
Bob left - I feel the later albums score much more highly compared to the
early albums than they would from an "unbiased" panel.  There's also the
simple fact that we all have Internet access, which makes us untypical.)

Given that, 250 would probably be enough to give a solid enough result on a
simple "yes/no" question (as long as it wasn't closer than about 60/40), but
on picking other favourite bands, well unless there's a clear and obvious
winner (like the same single band listed by over 30% of respondents), it's
statistically a waste of time.  And basing conclusions on the dozen or so who
rose to the "top ten" bait really tells you nothing about anything except
those individuals.

- Andy

Andy Gilham/Andy_Gilham at msn.com/Andy_Gilham at compuserve.com
http://ourworld.compuserve.com/homepages/Andy_Gilham