Consensus values for NIST biological and environmental Standard Reference Materials |
| |
Authors: | I Roelandts E S Gladney |
| |
Institution: | (1) Department of Geology, Petrology and Geochemistry, University of Liège, B-4000 Sart Tilman, Liège 1, Belgium, BE;(2) Air Quality Group, ESH-17, MS J-978, Los Alamos National Laboratory, Los Alamos, NM 87545, USA, US |
| |
Abstract: | The National Institute of Standards and Technology (NIST, formerly the National Bureau of Standards or NBS) has produced
numerous Standard Reference Materials (SRM) for use in biological and environmental analytical chemistry. The value listed
on the “NIST Certificate of Analysis” is the present best estimate of the “true” concentration of that element and is not
expected to deviate from that concentration by more than the stated uncertainty. However, NIST does not certify the elemental
concentration of every constituent and the number of elements reported in the NIST programs tends to be limited.
Numerous analysts have published concentration data on these reference materials. Major journals in analytical chemistry,
books, proceedings and “technical reports” have been surveyed to collect these available literature values. A standard statistical
approach has been employed to evaluate the compiled data. Our methodology has been developed in a series of previous papers.
Some subjective criteria are first used to reject aberrant data. Following these eliminations, an initial arithmetic mean
and standard deviation (S.D.) are computed from remaining data for each element. All data now outside two S.D. from the initial
mean are dropped and a second mean and S.D. recalculated. These final means and associated S.D. are reported as “consensus
values” in our tables.
Received: 25 April 1997 / Revised: 24 July 1977 / Accepted: 25 July 1997 |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|