On Tue, 12 Aug 2003, Torsten Franz asked:

> as a basis for Multidimensional Scaling I use different matrices: a
> distance matrix and a correlation matrix (Spearman). The results are
> different, too.  Why?  I use STATISTICA.

Mainly because a correlation matrix is NOT a distance matrix, I should
think.
 A distance matrix has 0 on the major diagonal and values elsewhere that
get larger the more distant each pair of objects is;  a correlation
matrix has 1 on the major diagonal and values that get smaller the more
distant each pair of objects is.  Why would one not expect different
results?

 -----------------------------------------------------------------------
 Donald F. Burrill                                         [EMAIL PROTECTED]
 56 Sebbins Pond Drive, Bedford, NH 03110                 (603) 626-0816

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to