[R] power.t.test function

Claus Ekstrøm ekstrom at dina.kvl.dk
Wed Feb 19 00:48:21 CET 2003

> Andrew.Swan at csiro.au writes:
> > Hi - Is it possible to use the power.t.test function in a scenario where the
> > number of observations in each group is different?
> No. It's not rocket science to modify the p.body expression, though.
> The tricky bit is what to do instead of "solve for n".

One possibility would be to have a fixed number >= 1 as the (constant) ratio between the two
group sizes. That shouldn't be too hard to implement. Next step is different variances ....


Claus Thorn Ekstrøm <ekstrom at dina.kvl.dk>
Dept of Mathematics and Physics, KVL
Thorvaldsensvej 40
DK-1871 Frederiksberg C
Phone:[+45] 3528 2341
Fax:  [+45] 3528 2350

More information about the R-help mailing list