Date: Thu, 15 Nov 2001 15:32:17 -0500
Reply-To: "William B. Ware" <wbware@EMAIL.UNC.EDU>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: "William B. Ware" <wbware@EMAIL.UNC.EDU>
Subject: Re: Negative correlation between factors
Content-Type: TEXT/PLAIN; charset=US-ASCII
The "valence" or sign of the factor is somewhat arbitrary. What you
describe is not at all unusual... just reverse the negatives to positives
and everything will be consistent...
William B. Ware, Professor and Chair Educational Psychology,
CB# 3500 Measurement, and Evaluation
University of North Carolina PHONE (919)-962-7848
Chapel Hill, NC 27599-3500 FAX: (919)-962-1533
http://www.unc.edu/~wbware/ EMAIL: firstname.lastname@example.org
On Thu, 15 Nov 2001, Catherine Guertin wrote:
> Dear all, can anyone help me with that?
> I performed a factor analysis on 14 items (unweigthed least squares method
> with oblimin rotation) and got 3 clearly distinguishable factors without any
> cross loadings.
> What I do not understand is that the loadings of every items on the second
> factor are negative (even though I did not forget to recode any of them).
> Accordingly, the factorial correlations between this factor and the 2 others
> are negative. The thing is that when I look at the pearson correlations
> between the 3 variables that I computed by calculating the mean of the items
> composing each factor, they are all positive (as they should be,
> theoretically speaking).
> How would you explain this difference between the factorial and the pearson
> correlations(positively versus negatively correlated)? How should I
> interpret the negative loadings of the factor analysis?
> Many thanks
> Catherine Guertin
> University of Montreal
> Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp