Re: Death to ABXers! Read Testament Here!

"ScottW" <ScottW48@[EMAIL PROTECTED]
> wrote in message 
news:d2e1879a-f457-4e31-9159-877006388552@[EMAIL PROTECTED]
 May 4, 12:00 pm, "Soundhaspriority" <nowh...@[EMAIL PROTECTED]
> wrote:
> "ScottW" <Scott...@[EMAIL PROTECTED]
> wrote in message
>
>
news:fcecf7b2-33cf-4de7-b04d-3ce278c792a9@[EMAIL PROTECTED]
> On May 4, 7:53 am, "Soundhaspriority" <nowh...@[EMAIL PROTECTED]
> wrote:
>
> > I picked this up on rec.audio.pro. Thank Heavens the ABXers will Burn
in
> > Hell in Torment Everlasting!
>
> Except no one ever said you have to listen to A then B then X.
> In fact one should listen A - X then B - X repeat as many times a you
> think is required to decide....if you can.
>
> However your reference does indicate how break-in is largely a human
> factor and not an equipment factor.
>
> ScottW
>
---------------------------------------------------------------------------­------------------
> You made two points.
>
> 1. The problem which Robert Adams discovered is that the trials are not
> independent. This throws all statistics out the window.

No it doesn't.  Just deal with it.  Do an A-X comparison.  Then do a B
listen to wipe A before a B-X, then an A listen to clear B before
repeating A-X.

[snip]

Fine for personal use. I have no objection. I would use it, if it were 
available. On the other hand, the ABXers are very statistics conscious, 
citing "statistically meaningful" thresholds of probability, etc. Adams's 
observation punches a big hole in that. It makes the results
mathematically 
incorrect. The thresholds are skewed in unknown ways. One can't be a
partial 
mathematical purist. When doing statistical research, you either do it 
right, or you're just a ****up. There is a difference between "personally 
meaningful" and "publishable." In the latter, one takes responsibility for

the welfare of others.

Bob Morein
(310) 237-6511