Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: PCA and rotation


From   kornbrot <d.e.kornbrot@herts.ac.uk>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: PCA and rotation
Date   Mon, 28 Dec 2009 11:41:22 +0000

Title: Re: st: PCA and rotation
The following intereesting article recommends
  1. ALWAYS use factor analysis not principal components, as errors are included in PC anf may differ across replications
  2. ALWAYS use oblique rotation rather than orthogonal rotation, as otherwise you may miss higher order factors
Reeve, C. L., & Blacksmith, N. (2009). Identifying g: A review of current factor analytic practices in the science of mental abilities. Intelligence, 37(5), 487-494. http://www.sciencedirect.com/science/article/B6W4M-4WN8H8G-1/2/b92c4e9a744cb75285467c53e906aed6.

Makes sens to me.
Other views?

Best

diana

On 21/12/2009 21:33, "Michael I. Lichter" <mlichter@buffalo.edu> wrote:

I recently found that when I extracted components using -pca-, rotated
them using an orthogonal rotation (e.g., -rotate, varimax-), and scored
them using -predict-, the correlations between what I presumed were
uncorrelated factors were actually as high as 0.6. I know that component
scores may be correlated, but this seemed a bit much. Somebody else
noted the same thing a few months ago
(http://www.stata.com/statalist/archive/2009-08/msg00793.html). On the
other hand, I found that factor scores (produced with -factor, pcf-) for
the same data remained virtually uncorrelated after orthogonal rotation.

I therefore assumed that the behavior of rotated PCs was a bug. I
contacted Stata. Isabel Canette told me that I was mistaken. She
referred me to "Methods of Multivariate Analysis" by A. Rencher, Second
Edition,Wiley, 2002, page 403, where Rencher says:

  "...If the resulting components do not have satisfactory interpretation,
   they can be further rotated, seeking dimensions in which many of the
  coefficients of the linear combination and near zero to simplify
  interpretation.
    However, the new rotated components are correlated, and they do not
  successively account for maximum variance. They are, therefore, no
  longer principal components in the usual sense, and their routine use
  is questionable".

In other words, it's not a bug, it's ... something else. Isabel said
that for this reason Stata discourages the use of rotation after -pca-.

What's odd is that I've seen a number of articles that use varimax
rotations  (with Kaiser normalization) of principal components in scale
development. The authors only use the PCA to guide scale development;
they perform further analysis with Cronbach's alpha and create summative
scales rather than using factor scores. Still, their interpretation of
the components are based on rotated component loadings that, at least
from Rencher's perspective, are "questionable".

--
Michael I. Lichter, Ph.D. <mlichter@buffalo.edu>
Research Assistant Professor & NRSA Fellow
UB Department of Family Medicine / Primary Care Research Institute
UB Clinical Center, 462 Grider Street, Buffalo, NY 14215
Office: CC 126 / Phone: 716-898-4751 / FAX: 716-898-3536

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



Professor Diana Kornbrot
email: 
d.e.kornbrot@herts.ac.uk    
web:    http://web.me.com/kornbrot/KornbrotHome.html
Work
School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK
 voice:   +44 (0) 170 728 4626
   fax:     +44 (0) 170 728 5073
Home
 
19 Elmhurst Avenue
 London N2 0LT, UK
    voice:   +44 (0) 208 883  3657
    mobile: +44 (0)
796 890 2102
   fax:      +44 (0) 870 706 4997








© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index