Statistical considerations for a parallel trial
where the outcome is a measurement.

Significance Level -- sided (default= .05, 2 sided).

Standard Deviation of the outcome variable, if known.

Enter two of the following three values and remaining value will be calculated.

  1. Power, usually .8 or .9.
  2. Total number of patients.
  3. Minimal detectable difference(Specify one of the following):
    1. Difference in means.
    2. % Location of the mean of one treatment group in terms of a percentile of the other treatment group.
Last Updated Tue Oct 24 22:56:39 EDT 1995
David A. Schoenfeld, Ph.D.,
With support from the Massachusetts General Hospital Mallinckrodt General Clinical Research Center, Research Resources Division, National Institutes of Health, General Clinical Research Center Program.