<- file 95mds.html ->

... MDS, unfolding, Multi-dimensional scaling?

=====================Forrest Young, 16 Apr 1995========sse Message-ID: <v01530500ad994e957780@[152.2.85.134]> From: forrest@unc.edu (Forrest Young) Subject: Re: Coombs unfolding model >I am looking for freeware programs in unfolding analysis, for my PhD >thesis. Does anyone know about it? >Any (little) indication would be greatly appreciated since this data >analysis method seems not to be widely used. >Thanks a lot in advance. Unfolding can be done by most of the multidimensional scaling programs. I know that the SAS MDS procedure and the SPSS ALSCAL procedure (of which I am the author) can do unfolding. However, unfolding is well documented as having frequent problems of "degeneracy" --- situations in which the analysis results in a solution which is a mathematical artifact that doesn't tell you much (if anything) about the data being analyzed. The artifact is evidenced by a collapse of many of the "row points" into a single point with the "column points" arranged in a circle whose center is the row points. I usually recommend that Principal Components be used instead, perhaps with the assumption that the data are at the ordinal level of measurement. This can be done with the SAS PRINQUAL procedure. I believe there is also an SPSS procedure, but I'm not sure which one it is. * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Document by Rich Ulrich. E-mail to wpilib+@pitt.edu

FAQ top.

Ulrich home page.

Ulrich FAQ. http://www.pitt.edu/~wpilib/stats99.html