Date:         Wed, 10 Mar 1999 07:21:13 -0500
Reply-To:     Jesus Olivero <oliverov@PILOT.MSU.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Jesus Olivero <oliverov@PILOT.MSU.EDU>
Subject:      Dr. Dennis Fischer
Content-Type: text/plain; charset="us-ascii"

Dear Dr. Fisher. I saw your name in internet as an expert in Cluster
Analysis and I wonder if you could help me with this question.
 
I am analizing Mercury in sediments from different locations and different
times.
 
My data looks like this:
 
        Station 1   Station 2    Station 3
 
Jan       0.15        0.13          0.13
Feb       0.20        0.33          0.12
March     0.19        0.22          0.21
 
I did a Cluster analysis using euclidean distance between groups and i get
an interesting cluster in which it was clear the relationships between the
different stations. When i submitted the paper for publication one referee
mentioned i could not use cluster analysis because my only variable was
Mercury in sediments. However, i believe my results were interesting in
terms of predicting what i was expecting!!
 
Is the cluster analysis restricted only to multivariate approaches?
 
Thanks a lot for your time and collaboration,
 
 
Best regards,
 
 
Jesus Olivero
Department of Pharmacology and Toxicology
Michigan State University
East Lansing, MI 48824