Date:         Mon, 29 Apr 2002 12:49:17 -0700
Reply-To:     "Dennis G. Fisher" <dfisher@CSULB.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Dennis G. Fisher" <dfisher@CSULB.EDU>
Subject:      Re: cluster analysis
Comments: To: nikhil-monika taparia <nikmon@HOTMAIL.COM>
Content-Type: multipart/alternative;

Yes, you could use ANOVA to assess for differences among the clusters,
but you would always do it on variables that were NOT included as input
to the cluster analysis.  These other variables are called "naturalness"
variables in cluster analysis lingo.  This problem is discussed on pages
64-65 of Aldenderfer & Blashfield's (1984) book by SAGE on Cluster
Analysis.
Dennis Fisher

nikhil-monika taparia wrote:

> Hi, I did cluster anaysis on a data set and came up with 4 clusters of
> employees based on their 16 'work styles' variables.  When I do
> profiling of these clusters on their demographic information, I do not
> see much of a difference among the clusters.  So. I attempted to do
> PROC ANOVA among the clusters using the demographic variables.
> Rightly, hardly on two of the 15 demographic variables I used, Schefe
> Test shows any difference in the clusters. My queries: 1.  Is it
> correct to try profiling the clusters on their demographic information
> or should one do it on the 16 'work style' variables? 2. Am I right in
> using the PROC ANOVA to see the differences among the clusters? This
> is for an assignment I got in class.  I am not cheating by posing this
> query here coz I am tryiong to do more than what the assignment ask us
> to do.  The above mentioned queries are NOT part of the
> assignement. Expert help appreciated.  Thanks. Nikhil Taparia
>
>
> -----------------------------------------------------------------------
> Get your FREE download of MSN Explorer at http://explorer.msn.com.

--
Dennis G. Fisher, Ph.D.
Director
Center for Behavioral Research and Services
1090 Atlantic Avenue
Long Beach, CA 90813
562-495-2330
562-983-1421 fax

To him who devotes his life to science, nothing can give more happiness
than increasing the number of discoveries.  But his cup of joy is full
when the
results of his studies immediately find practical application.  There
are not two sciences.  There is only one science and the application of
science,
and these two activities are linked as the fruit is to the tree.
   --Louis Pasteur


[text/html]