From: J. Wayde Allen (wallen@lug.boulder.co.us)
Date: 11/06/00-06:02:36 PM Z
On Mon, 6 Nov 2000 bmaxey1@juno.com wrote:
> Testing all the variables in a typical formula for a simple emulsion is a
> daunting task. Your task becomes easier as you learn more about what each
> component in the formula does or should do. Without boring the list, I
> can easily come up with hundreds of variables that can cause changes.
Yes!!! However, with one-variable-at-a-time testing you only get a single
data point no matter how many variables you used for the test. If you
change from this condition all bets are now off. If you use multivariate
testing methods you do basically the same amount of work in the lab, but
at least get some indication about the multidimensional response
surface. That allows you to make some predictions about what happens if
more than one variable changes. The whole idea is to get more information
for the same amount of work.
- Wayde
(wallen@lug.boulder.co.us)
This archive was generated by hypermail 2b30 : 12/01/00-11:46:56 AM Z CST