Equivalence testing for use in psychosocial and services research: An introduction with examples

This paper provides an introductory presentation of current equivalence methodology that is a preferred alternative to usual power analyses when one wishes to assert that performances on given outcome measures for different samples are equivalent, that is, could have come from the same population. Such an approach may be important for mental health policy makers who are comparing effectiveness of two systems of care that appear to have similar outcomes but nontrivial differences in cost or other organizational characteristics. Copyright (C) 1996 Elsevier Science Ltd