ISO  5725-l :1 994 pdf download – Accuracy (trueness and precision) of measurement methods and results – Part 1: General principles and definitions.
3.5 accepted reference value: A value that serves as an agreed-upon reference for comparison, and which is derived as: a) a theoretical or established value， based on scientific principles; b) an assigned or certified value, based on exper- imental work of some national or international or- ganization; c) a consensus or certified value, based on collabor- ative experimental work under the auspices of a scientific or engineering group; d) when a), b) and c) are not available, the expec- tation of the (measurable) quantity, i.e. the mean of a specified population of measurements. [ISO 3534-1] 3.6 accuracy: The closeness of agreement between a test result and the accepted reference value. NOTE 2 The term accuracy, when applied to a sct of test results, involves a combination of random components and a common systematic error or bias component. [ISO 3534-1]
19 The definitions given in 3.16 and 3.20 apply to results that vary on a continuous scale. If the test result is discrete or rounded off, the repeatability limit and the reproducibility limit as defined above are each the minimum value equal to or below which the absolute difference between two single test results is expected to lie with a probability of not less than 95 %. 20 The definitions given in 3.8 to 3.11, 3.15, 3.16, 3.19 and 3.20 refer to theoretical values which in reality remain un- known. The values for reproducibility and repeatability stan- dard deviations and bias actually determined by experiment (as described in ISO 5725-2 and ISO 5725-4) are, in stat- istical terms, estimates of these values, and as such are subject to errors. Consequently, for example, the probability levels associated with the limits r and R will not be exactly 95 %. They will approximate to 95 % when many labora- tories have taken part in the precision experiment, but may be considerably different from 95 % when fewer than 30 laboratories have participated. This is unavoidable but does not seriously detract from their practical utility as they are primarily designed to serve as tools for judging whether the difference between results could be ascribed to random uncertainties inherent in the measurement method or not. Differences larger than the repeatability limit r or the repro- ducibility limit R are suspect. 21 The symbols r and R are already in general use for other purposes; in ISO 3534-1 r is recommended for the correlation coefficient and R (or W) for the range of a single series of observations. However, there should be no COI- fusion if the full wordings repeatability limit r and reproduc- ibility limit R are used whenever there is a possibility of misunderstanding. particularly when they are quoted in standards.