>-----Original Message----- >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] On Behalf Of Hank Stevens >Sent: Monday, March 31, 2003 2:43 PM >To: [EMAIL PROTECTED] >Subject: [R] power.t.test > > >power.t.test() requires a standard deviation as input, but for >a two sample >test, I do not understand what is required. Is it the standard >deviation of >the means (standard error), or of the combined sample? > >thanks, >Hank Stevens
The 'sd' argument is the standard deviation of the mean for *each* sample, keeping in mind that the underlying assumption is that both samples have equal variances. Thus you need only specify one value. HTH, Marc Schwartz ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help