Dimitris
Take a look on MDS-X.. http://newmdsx.com/library.htm Best Naji -----Message d'origine----- De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] la part de Dimitis Vatikiotis Envoy� : jeudi 29 janvier 2004 17:15 � : [EMAIL PROTECTED] Objet : [edstat] how to perform multidimensional unfolding? for example: a dataset of 100 respondents containing the following variables a) age (4 age groups) b) area (5 area categories) c) educ (3 education levels) d) 15 attributes (5-point scale) Using any common statistical software it is rather simple to perform classical mds and create a 2-dimensional map for the 15 attributes. What I cannot do is to find the position of age, area and educ in that map (preserving the initial coordinates for the 15 attributes) , i.e. to create a joint-plot of attributes and categories for (a), (b), (c) I read that this is called multidimensional unfolding. Could anyone recommend me an appropriate software (I've already tried SPSS) and/or any useful reading (preferably something that includes examples). I would very much appreciate any help. Dimitris . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . ================================================================= . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
