Description
REPEALED Feb 2025. The procedure shows how to estimate measurement uncertainty based on internal reproducibility data, and how to check that your estimate is adequate by participation in proficiency testing schemes. The aim was to make a simple and user-friendly guideline by using real examples, in order to get laboratories started on conducting estimations of measurement uncertainties. The procedure also shows how to use data from collaborative studies for estimation of the uncertainty, which presupposes that the laboratory verifies the obtained precision of the study. Further, the procedure describes the relation between standard deviations and concentration levels based on a number of parallels from collaborative studies, which in turn indicates the precision expected of the laboratory.
An Excel spreadsheet illustrating estimation of measurement uncertainty is available for download here.
21.05.24: The spreadsheet is correct and based on “6.2 Estimation of measurement uncertainty based on internal reproducibility from different matrixes and levels at the same time”. Unfortunately, typos in the procedure itself on page 23 says Sr = 0.042, which should be Sr = 0.046, and says Sx = 0.035 which should be Sx = 0.038.)
06.02.25: The NMKL Procedure 8 has been withdrawn and will no longer be supported. The reason for withdrawal was that the procedure needed revision, and NMKL found that other standardization organizations already have such procedures of high standard.