Hi,
Could someone clarify the use of the -v smoothing option.
On the website, under one sample T-test, it says: "If you have fewer
than 20 subjects (approx. 20 DF), then you will usually see an
increase in power by using variance smoothing".
Firstly, is it appropriate to use? Many recent threads in the
archives say you shouldn't use it.
Secondly, is this rule of thumb applicable for two sample T-tests, if
the combined group size is below 20?
Thanks
Suzanna
|