Re: Larry Athy and His Logical Analyses.

"Larry Athy" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> On Mon, Dec 1, 2003, 12:04pm (CST+6), JMB posted quite a bit here and
> only succeeded in demonstrating that he does not have the ability to
> understand the various methods of interpreting the different types of
> problems covered by Williams.
> -  JMB repeatedly refers to a method used to interpret "p" in a problem
> set up to compare performance results with an accepted standard.  An
> acceptable "p", such as p < .05 would indicate acceptable performance,
> while a larger "p" would indicate unacceptable performance.
> -  My data can not be interpreted by this method.  I am not trying to
> determine if anything meets an accepted standard.  I am trying to
> determine how closely related two groups of data might be.
> -  Williams explains both methods, and others, in the copy of his book
> that I have.
> -   JMB simply does not understand that there are different types of
> relationships to be analyzed from calculations done in the same manner,
> except for final interpretation.

This is getting boring.  Appart from the irrelevance due to the fact that
even if the techniques were applied correctly they cannot do what Larry
claims for them, it is clear that Larry is opperating in a fantasy world.
The closest thing in the Williams book to what Larry claims is the method he
used is the bottom paragraph of table 11.1 is which he explains:
"t is interpreted in Table 7.2, using d.f. = n-2.  given this
interpretation, the r is +.937 would be expected to occur in fewer than one
of 100 cases if due to sampling error (that is, p < .01)"
Table 7.2 is an example of a Table of t.  It does not say, nor does Williams
claim that it means, that there is over 99% correlation between the sets.
That is something Larry has made up, either through ignorance of what a
t-table gives or through an intent to deceive.

>           Regards,  Larry Athy, P.E.
>
>
>
>

-- 
John Byrne
www.iol.ie/~archaeology
To email me use the feedback form on the website.
The address attached to this post is just a spam trap.