From: Jeffrey D. Mathias (jeffrey.d.mathias@worldnet.att.net)
Date: 11/06/00-11:24:13 AM Z
"J. Wayde Allen" wrote:
> ...
> In a process with highly correlated variables, this is true, and is
> exactly why one-variable-at-a-time testing is NOT a good approach. You
> can never see variable correlation with this kind of testing, and as such
> changing anything throws everything off. You don't have to take my word
> for it though see <http://www.pdlab.com/experimentx.htm>.
> ...
Wayde,
As a friendly jester and for the sake of some professionalism on a
professional topic:
If we are to talk good science, let us also rely on good scientific
information. The pdlab site you reference does not seem to divulge it's
author(s) anywhere. I believe that it is good practice that any
scientific source of information present it's author(s) and necessary
resources/bibliographies. Otherwise, why should we take anyone's word
for it. One can find just about anything on the internet these days.
I believe your argument would hold more credibility with a credential
bearing reference.
-- Jeffrey D. Mathias http://home.att.net/~jeffrey.d.mathias/
This archive was generated by hypermail 2b30 : 12/01/00-11:46:56 AM Z CST