|Date: ||Fri, 14 Jul 2000 11:28:42 -0500|
|Reply-To: ||"Nichols, David" <nichols@SPSS.COM>|
|Sender: ||"SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>|
|From: ||"Nichols, David" <nichols@SPSS.COM>|
|Subject: ||Re: cutoff scores for Discriminant?|
The SPSS DISCRIMINANT output does not include cutoffs. In the case of two
groups, it's not that hard to figure it out in terms of the single canonical
discriminant function (if equal priors are used, just take the point half
way between the two group centroids), but if there are two or more
functions, it gets very complicated, as you'd have to describe lines cutting
off areas in a two or higher dimensional space.
Principal Support Statistician and
Manager of Statistical Support
> -----Original Message-----
> From: Lary Jones [mailto:ljones@BINGHAMTON.EDU]
> Sent: Thursday, May 04, 2000 4:43 PM
> To: SPSSX-L@LISTSERV.UGA.EDU
> Subject: [SPSSX-L] cutoff scores for Discriminant?
> I have a user who has asked for cutoff scores for DISCRIMINANT. Any
> suggestions out there?
> Lary Jones % Statistical Computing Analyst
> Computing Services % ..........................
> Binghamton University % LJones@Binghamton.EDU
> Binghamton, NY 13902-6000 % (607) 777-2614