A quantitative term to describe the (lack of) accuracy of a
chemical measurement process; comprises the
imprecision and the
bias. Inaccuracy must be viewed as a two-component quantity (vector); imprecision and
bias should never be combined to give a scalar measure for chemical measurement process
inaccuracy. (One or the other component may, however, be negligible under certain
circumstances.) Inaccuracy should not be confused with uncertainty. Inaccuracy (imprecision,
bias) is characteristic of the measurement process, whereas error and uncertainty are
characteristics of a result. (The latter characteristic, of course, derives from the
imprecision and bounds for
bias of the chemical measurement process.)
Note:
The resultant
bias and imprecision for the overall measurement process generally arise from several
individual components, some of which act multiplicatively (e.g.
sensitivity),
and some of which act additively (e.g. the blank).
Source:
PAC, 1995, 67, 1699
(Nomenclature in evaluation of analytical methods including detection and quantification
capabilities (IUPAC Recommendations 1995))
on page 1706
PAC, 1989, 61, 1657
(Nomenclature for automated and mechanised analysis (Recommendations 1989))
on page 1663