|Date: ||Sat, 27 Apr 1996 14:32:07 -0500|
|Sender: ||"SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>|
|From: ||Kerry Patrick Meline <kpmeline@STUDENTS.WISC.EDU>|
|Organization: ||University of Wisconsin-Madison|
|Subject: ||pairwise in cluster analysis|
SPSS doesn't have an option for pairwise deletion of missing data for
hierchial clustering (only listwise or include), but this is available
in k-means clustering (pairwise, listwise, or include).
Is this because of the algorithms used in hierchial?
Any ideas about getting around this data loss with binary (0,1)
data...some method of imputation?