Estimating the reliability of a method
Why carry out interlaboratory studies?

Fidelity evaluates the difference between repetitions made under specified conditions on the same sample. It is generally expressed numerically by characteristics such as standard deviation, variance or coefficient of variation. It is these specified conditions that are variable and lead to the estimation of several fidelities. For example, if two laboratories repeat the same method on the same sample, they will of course find different measurements from which at least three estimates can be calculated: overall from all measurements, with the measurements of the first laboratory only, and with those of the second.

The figure "Main sources of variation influencing the precision of a method" presents a diagram...