I'm interested in comparing frequencies among groups which have
ordering. The data are composed of two-way contengency tables; the row
variable having 2 or more levels and the column variables having 2
levels. I'd like to see what the effect of testing for trend has on the
p-values as compared to a simple Chi-squared analysis ignoring trend.
Has anyone had the occasion to perform such an analysis in Splus and has
accompanying S-code? Any help would be appreciated.
--
,,,
(o o) Mohamed Abdolell
--oOO--(_)--OOo-------------------------------------------------------
Mount Sinai Hospital, Samuel Lunenfeld Research Institute
850-600 University Ave. Toronto, Ontario, M5G 1X5, CANADA
e-mail: abdolell@mshri.on.ca voice: (416)586-8886 fax: (416)586-8404
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|