Hi Ms. Scheltema!
The formula depends on whether you have equal variances for the
covariate and the outcome variable. if so you can use:
d = (Ybar(Experimental Group) - Ybar(Control Group))/sqroot(std.dev for
Y times(1 - squared correlation of X and Y)).
If these variances are not equal, the formula becomes rather more
complicated.
Hope this helps.
Gary Winkel
City University of New York
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
