[R] PCA with random effects

[David Sidhu]

I have the results of a rating study in which ~30 participants rated a subset
of 20 items on 25 different dimensions.

I would like to perform PCA on these ratings to reduce the 25 different
dimensions. However, instead of doing this on the mean ratings for each
item, I would like to perform the PCA at the level of individual ratings. I
am assuming that in order to do this I have to specify the fact that ratings
are nested within participants and within items (i.e., crossed random
effects for subjects and items). Is this possible?

---
David M. Sidhu, MSc<http://davidmsidhu.com/>
PhD Candidate
Department of Psychology
University of Calgary







	[[alternative HTML version deleted]]