On 28 May 2000 17:25:18 GMT, [EMAIL PROTECTED] (Wen-Feng
Hsiao) wrote:
Dear Ulrich,
Thanks for your reply. I now have more confidence in analyzing my data
with PCA. However, you mentioned that 43 subjects are too thin to obtain
a stable/converge result. This incurs me another question --
Dear listers,
I have 43 subjects to rank 11 stimuli. To obtain the underlying
variables, principal components, I conduct a PCA (Principal Component
Analysis) to the obtained ranked-data. However, a friend of mine told me
that using ranking data to do a PCA is quite dangerous, since the scores
