Dear all,
I have a large set (n=500) of non-normal data which take only nineteen
discrete values. I've been attempting various transformations to achieve
approximate normality.
My query is this:
a) Is the Shapiro-Wilk test appropriate for this situation?
b) does anybody have a good reference?
Thanks,
Jim Lewsey
====================================================================
James D Lewsey BSc(Hons)
Research Fellow
National Centre for Transcultural Oral Health
Eastman Dental Institute for Oral Health Care Sciences
University of London
256 Gray's Inn Road
London WC1X 8LD
Telephone(Direct Line) +44 (0) 171 915 1192
Department Fax +44 (0) 171 915 1233
Department Telephone +44 (0) 171 915 1193
Eastman Website http://www.eastman.ucl.ac.uk
====================================================================
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|