Question on precision and related

[JI2002]

http://chromatographyonline.findanalyti ... ail/747883

I read this on LCGC North America and am really confused when precision and error are used sort of interchangeably in the article. I always thought you can have high precision (low variability) for a measurement even the error is high (low accuracy). The second question is, is the square of coefficient of variation (CV) additive? I always thought the variance σ2 is additive, but not sure about CV square. Hope someone can shed some light on this.

[tom jupille]

I always thought you can have high precision (low variability) for a measurement even the error is high (low accuracy).

"Precision" and "accuracy" are well-defined terms. "Error" is not and has to be taken in context. Since the entire article deals with precision problems, I didn't see any ambiguity.

I always thought the variance σ2 is additive, but not sure about CV square.

CV is σ divided by the mean. Whenever you add together a series of fractions having the same denominator, the denominator stays constant (you simply add the numerators). Therefore yes, CV^2 is additive.