YetAnotherForum
Welcome Guest Search | Active Topics | Log In | Register

Correspondence Analysis Question Options · View
Previous Topic · Next Topic
brm42
#1 Posted : Thursday, October 15, 2009 5:59:28 AM
Rank: Newbie

Groups: Premium , Registered
Joined: 10/15/2009
Posts: 3
When viewing the results of a correspondence analysis, I have several variables that are stacked on top of one another. Is it okay if I combine these variables? If they are conceptually the same, should I combine them into one category? Or should I continue to treat them separately?
Back to top
 Report Abusive | Report SPAM
Sponsor  
 
Back to top  
rsilva
#2 Posted : Thursday, October 15, 2009 7:00:03 AM
Rank: Member

Groups: Premium , Registered
Joined: 6/11/2009
Posts: 19
Are you using a 2D view? I suspect there is a 3rd dimension to the solution
Back to top
 Report Abusive | Report SPAM
brm42
#3 Posted : Thursday, October 15, 2009 7:26:40 AM
Rank: Newbie

Groups: Premium , Registered
Joined: 10/15/2009
Posts: 3
Oh right... good call! Thanks for the help!
Back to top
 Report Abusive | Report SPAM
manandarajan001
#4 Posted : Thursday, October 15, 2009 8:02:33 AM
Rank: Advanced Member

Groups: Administrators , Premium, Registered, WebAdmin
Joined: 5/29/2009
Posts: 105
Yes, I would not collapse it, but rather think of it as a rubic cube. I suggest looking at the Eigen values and determine if there is indeed more than 2 dimensions. Then plot d1 vs d2; d2vs d3; and d1 vs d3. This should tell you the story
Back to top
 Report Abusive | Report SPAM
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

YAFPro Theme Created by Jaben Cargman (Tiny Gecko)
Powered by YAF | YAF © 2003-2008, Yet Another Forum.NET
This page was generated in 0.041 seconds.