C. Hartmann et al., REAPPRAISAL OF HYPOTHESIS-TESTING FOR METHOD VALIDATION - DETECTION OF SYSTEMATIC-ERROR BY COMPARING THE MEANS OF 2 METHODS OR OF 2 LABORATORIES, Analytical chemistry, 67(24), 1995, pp. 4491-4499
The usual way of testing the equivalence of two measurement procedures
is analyzed, and possible alternatives are discussed. It is concluded
that for method validation purposes, defining an interval of acceptab
le values and applying interval hypothesis tests is to be preferred ov
er the usually applied point hypothesis tests. Statistically testing a
gainst acceptance limits has the advantage that the beta-error of the
usual null hypothesis (there is no bias) can be limited. It is shown t
hat significance testing at a fixed beta-error requires, however, reve
rsal of the usual null and alternative hypotheses. Equations are prese
nted for the case in which the acceptable bias is expressed as a ratio
of the reference mean. Moreover, the principle of randomization tests
is explained, and their possible advantages over the ordinary test st
atistics in method validation are discussed.