For a two-tailed test with type I error, alp, with inputted mean and average group sd will have power equal to pow.
[TYPE INTO R THE DESIRED INPUTS ALPHA, MEAN, SD AND N USING VALUES IN FORM BELOW].
The R syntax below
pow <- 0.8 beta <- 0.05 mdiff <- 2 sd <- 4
the the below will give the equal samples in nout.
fn <- function(n) {
(pow - (1 - pt(abs(qt(beta/2,(2*n)-2)),(2*n)-2,mdiff/sqrt(2*sd*sd/n))) )^2
}
nout<- nlm(fn,2)
nout<- trunc(nout$estimate+1)
print(nout)This can be run in SPSS 16 using R integrator plug-in (see here).