What is method precision?

Precision, or more correctly the imprecision, of an analytical method is an expression of the expected degree of spread about the average value of replicate measurements made under prescribed conditions.
The precision is usually stated in terms of standard deviation (SD) or the relative standard deviation (%RSD). For a set of n values {x1, x2,…xn}SD and %RSD are calculated as follows:

The size of precision depends on a variety of factors, including the number of parameters that are varied during the precision study and the level of variation in the operation conditions. At Chemex, two types of precision are estimated: repeatability and reproducibility.

Repeatability is a type of precision relating to measurements made under repeatability conditions, i.e. same method; same material; same operator; same laboratory; narrow time period. This measure is a short-term variation in measurement results and is determined by analysing a minimum of 7 sample replicates in one analytical batch and instrument run, and is used to establish compliance with method performance criteria. On its own, it cannot indicate the spread of results that can be expected over the longer term.

Reproducibility is a type of precision relating to measurements made under reproducibility conditions, i.e. same method; same material; different operator; different equipment; even different laboratories (i.e. Proficiency testing schemes); long time period. At Chemex, this term is used for long term method performance such as AQC sample data generated over a longer period.

After initial method validation, an ongoing assessment of method precision will be checked by carrying out intermittent studies, where replicates of certified reference material (CRM) or spiked matrix samples are analysed in a single batch, and reviewing daily AQC sample performance data spread.