Re: Blind Cable Test at CES

On Thu, 24 Jan 2008 01:48:57 -0800, "JBorg, Jr."
<EddieM@xxxxxxxxxxxxxxxxxxx> wrote:

Oliver Costich wrote:
JBorg, Jr. wrote:
Oliver Costich wrote:



snip


Physics is models based on observation of events repeating, or as
David Hume calls it "bad habit". The level of "certainty" or
probability of error can be reduced to infinitesimal levels but
uncertainty is still there even if we don't behave like it is. The
less controllable experiments are, and the farther the "science"
deviates from having rigorous mathematical models, the less
certainty you have. In particular you seem to be willing to abandon
all use of statistics as to you the results don't provide certainty
(and they never do).


But clearly, such test obviously did not proved that subtle
differences did not exist, and as you said, disproving the subtle
difference that audiophiles hear is a certainty NOT in the realm of
Statistical analysis.

You seems to be saying still that statistic analysis or some other
method can be use, but stat analysis is not it.

What am I suppose to do, Mr. Costich ?

Go learn something about science and measurement, statistical design
and analysis. If test after test shows that the hypothesis that
listeners cannot identify cables, then most scientifically minded
rational people would conclude it's unlikely to be true.


But test after test shows that there is no well-designed test as it shows.


So if a test doesn't give the answer you want the design is bad.
Since test after test fails to give the answer you want there is no
well designed test. Is that your position?

Do you know by divine knowledge that no test exists or are you relying
on there not being one so far, in which case you are relying on
statistical reasoning to support your claim.
.

Relevant Pages

  • Re: Blind Cable Test at CES
    ... abandon all use of statistics as to you the results don't provide ... Statistical analysis. ... So if a test doesn't give the answer you want the design is bad. ... Do you know by divine knowledge that no test exists or are you relying ...
    (rec.audio.opinion)
  • Re: ANOVA QUESTION - Terminology
    ... Depending on design, ... When Richard Ulrich mentioned something he said is "offered ... in every course or lext in experimental design" and indelicately ... Daniel, and characterized him as "not a household name in statistics, ...
    (sci.stat.consult)
  • Re: POTM nomination: Nothing but a fairytale
    ... The Earth has no moon. ... propablilities only show how unfit you are to do statistics. ... statistical analysis. ... starting point to come to these specific sequences. ...
    (talk.origins)
  • Re: Randomized block design
    ... > I use a randomized block design with 4 treatments and a control. ... > interaction between time and Ulva present/absent and the interaction ... Department of Mathematics and Statistics ...
    (sci.stat.math)
  • Re: Statistical Pattern Recognition
    ... the recent brain research is the *evidence* that shows brains perform ... statistical analysis during pattern recognition? ... that it is an *associative memory* process, not statistical analysis. ... I've already provided several references dealing with the statistics ...
    (talk.origins)