February, 2003
© 2000, 2001, 2003 Scott S. Emerson, M.D., Ph.D.
Session 1- Fixed Sample Design: 22
Scientific Hypotheses: Primary Endpoint
As a general rule, the usual formulation of the hypotheses from the scientific standpoint does not lend itself to statistical analysis.
Further refinement of hypotheses often needed
·Endpoint modified to increase precision
·Statistical model to account for variation in response
·Precise statement of hypotheses to be discriminated
-Due to sampling variability, contiguous hypotheses cannot be discriminated with finite sample size