|Date: ||Wed, 24 Jan 2001 16:51:46 -0500|
|Reply-To: ||Arthur J Kendall <KendallA@GAO.GOV>|
|Sender: ||"SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>|
|From: ||Arthur J Kendall <KendallA@GAO.GOV>|
|Subject: ||Re: Gower's Coefficient|
Many proximity measures have multiple names. SPSS has about a couple dozen measures, perhaps one of them is the same or close to Gowers. Many of the measures have parameter that you can set and perhaps Gower's measure is one of the more general coefficients with particular parameter settings.
I suggest you go to http://www.pitt.edu/~csna/
and click on the discussion list CLASS-L.
This is the Classification Society of North America discussion list. If you were to post your question there, you might be able to find out if Gower's coefficient goes by some other name, or if some other coefficient would be more appropriate for your data.
Although John Gower is/was with the British Classification Society, he may be on that list.
In general, my feeling is that one should try a series of proximity measures in an analysis.
If you calculate the coefficient matrix outside SPSS, you can still read it into SPSS to do the clustering.
>>> "S. Zitzer" <sallyz@U.WASHINGTON.EDU> 01/24/01 04:14PM >>>
I have a user who needs Gower's Coefficient (a proximity measure).
I found a "non-guaranteed" SAS macro that will do it. Just wondering
if anyone has found it in SPSS, I've searched the manuals but don't
University of Washington, Seattle, WA, USA
****** Message from InterScan E-Mail VirusWall NT ******
** No virus found in attached file noname.htm
* All attachmes were scanned by Interscan 3.5, and no virusus were found.
***************** End of message ***************