ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

Proper Ways of Comparison of Two Laboratory Methods

J. Mocák, B. Balla, A. Bobrowski, and P. Blažíček

Department of Analytical Chemistry, Faculty of Chemical and Food Technology, Slovak University of Technology; SK-812 37 Bratislava

E-mail: mocak@cvt.stuba.sk

Received: 16 November 2002

Abstract: For statistically correct decision whether two laboratory methods provide concordant results when measuring the same objects, only those regression techniques are reliable, which respect the errors of both compared variables or are not influenced by them. Overview of such techniques is given in this paper. Ordinary least-squares regression may provide biased results leading to wrong decision making.

Full paper in Portable Document Format: 573a143.pdf

Chemical Papers 57 (3) 143–146 (2003)