[R] squared semi-partial correlation

N.Callow pes603 at bangor.ac.uk
Fri Aug 6 12:48:00 CEST 2004

Hi, can you help? I am really struggling to find the answer to the
following question.

I have conducted a standard multiple regression using SPSS. The dependent
variable is CE, and the independent variables are Ibelieve and Tbelieve.
Because the independent variables are highly correlated I have examined the
squared semi-partial correlations. I
understand that the Ibelieve squared semi partial correlation accounts for
a significant proportion of the variance in CE t= 2.689, whereas the
tbelieve does not t = .556. However, I want to see if the squared
semi-partial correlations are significantly different TO EACH OTHER
(Ibelieve .168, Tbelieve .035 N= 220). How do I do this?  I assume that I
will need to do this "by hand" with an equation? Also, do I have to take
into account that the squared semi-partial correlations are correlated in
terms of the dependent variable? 
Thanks very much.

Dr Nichola Callow (C.Psychol.)
School of Sport, Health, and Exercise Science
University of Wales, Bangor
George Building
Holyhead Road
Gwynedd LL57 2PX
Tel: 44 (0) 1248 388243
Fax: 44 (0) 1248 371053
E-Mail: n.callow at bangor.ac.uk

More information about the R-help mailing list