A decision problem frequently occurring in metrology is that of compatibility of data obtained by two (or several) different laboratories, methods or instruments. Several statistical approaches to this conformity assessment problem are reviewed including the classical "equality of means" hypotheses tests. A new compliance hypothesis is formulated when one laboratory is a prestigious national metrology institute whose data may be given as a certified reference material (CRM) interval for the measurand (or a gold standard). The Kullback--Leibler information number is suggested as a directional (asymmetric) interchangeability index.
Pub Type: Journals
Compliance testing, Behrens-Fisher problem, coverage intervals, likelihood ratio test, interchangeability index, power function.