Sorry Jonathan - I don't have the ref. or the documentation relating
to the current (or proposed) IUPAC recommendations readily to hand,
and I'm too tied up at present to locate them. (I don't see them
referred to on the IUPAC website - see
http://www.chem.qmw.ac.uk/iupac/ - but I may have missed them. Or
perhaps they have not yet been published.
As far as I remember the nomenclature recommendations (or proposed
recommendations) came as a Blackwell's publication. But I recall
that they rejected the traditional IFCC and clinical chemists'
identification of accuracy with bias (which, inter alia, conflicts
which normal English usage and the OED definitions).
The Dictionary of Statistical Terms definition tells us that
"accuracy refers to closeness of an observed result to the true or
accepted value" - and of course an observed result may differ from
the true or accepted value because of either systematic or random
errors or both.
Apologies if I misinterpreted your report, but since it spoke of
'accuracy and precision' I inferred you identified accuracy with bias.
--
Roger Ekins
Molecular Endocrinology
University College London
|