February, 2003
© 2000, 2001  Scott S. Emerson, M.D., Ph.D.
Session 2: 144
Evaluation of Designs
Power curve
Probability of rejecting null for arbitrary alternatives
·Power under null: level of significance
·Power for specified alternative
·
Alternative rejected by design
·Alternative for which study has high power
S+SeqTrial defines
·Power curves for upper and lower boundaries
·Alternatives having specified power for each boundary