testing margin superiority margin Philippe Rocca-Serra adapted from SEVCO. If non-inferiority is inferred, based on a prespecified non-inferiority margin, then one can subsequently test for superiority using a prespecified margin of zero. If so, one should recognize that the null hypothesis (set of assumptions and optimal dataset for analysis) is different for the superiority test than for the non-inferiority test. It is not expected that one would use this term when expressing parameters for a hypothesis test because this parameter is always set to a value of zero. The superior margin is distinct from the effect size for power calculation and the minimal important difference. minimal clinically important difference A hypothesis testing margin of zero used in a study with a superiority goal. The superiority margin (also sometimes called the minimal clinically important difference) defines how much better one treatment must be to be considered meaningfully superior — not just statistically different. Joanne Dehnbostel, Ken Wilkins, Harold Lehmann, Airton Stein ready for release