Puntipa Wanitjirattikal, PhD Candidate
Department of Statistics
Western Michigan University
Abstract:
This study proposes a test for statistical equivalence of two
measurements. Typically, a new measurement process Y is compared
to an existing or standard measurement process X. We are
assuming that X and Y are measurements on the same scale.
The paired t-test may be used to check for significant difference between
(X,Y) pairs. However, the paired t-test is intended to detect
shift-type relationships of the form Y = X + d and may have
low power for scale-type relations of the form Y = gX.
We propose a test that has reasonable power to detect
either shift or scale-type relationships. Secondly, we propose
a bioequivalence testing approach to swap the hypotheses
so that statistical equivalence of the two measurements
is the alternative hypothesis and bears the burden of proof.
Rather than being the default conclusion in the absence of
sufficient evidence, we conclude “clinical equivalence”
only if there is evidence to support the claim that the
magnitude of disagreement between the two measurements
lies within specified limits.
---
All statistics graduate students are expected to attend.