On Wed, 10 May 2000 08:42:27 +0100, "HAideren" <[EMAIL PROTECTED]>
wrote:
> What interpretations can be made from bipolar factors (factors consisting of
> negative as well as positive loadings on items)? When calculating the factor
> score for bipolar factors, should the negative sign be ignored?
If you start with a set of all-positive correlations, then the next
factor derived after the first one will be "bipolar", and you have
to respect that, totally. The rest of them are likely to be bipolar,
too.
If it is convenient, you may want to multiply EVERY loading
on a factor by -1, because the direction is arbitrary.
But the positives are in the opposite direction of the negatives.
If they do not exist in terms of a set of correlations like that,
then it should be hard to talk about "bipolar factors."
Personal experience, with scales of "symptoms", doing
principal factoring, followed by Varimax rotation:
A bipolar factor in the rotated solution is unusual. It might
signify that I need to add some items and subtract
(or reverse-score) some others, or it might mean
that I don't have a good technical solution; in any case,
adding-and-subtracting those high-loading items is not
apt to make a very robust total score. I take this as a warning
that I may have too many variables for the N, and I should
test some smaller subsets. Having more than one or two
items with "split-loadings" bothers me the same way.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
