Influence of method validation parameters in the measurement uncertainty estimation by experimental approaches in food preservatives analysis.

The statistical method of random uncertainties has been adopted for the uncertainty calculation of chemical measurements, since it is a good estimator of the overall uncertainty. Type B (inherited) uncertainties are usually insignificant when compared to type A (experimental) uncertainties. In quantitative measures, the mean is the most probable value of the measurand and the standard deviation of the mean is the best estimate of its standard uncertainty. Measurement uncertainty of qualitative measures may be estimated by the probability of taking right or wrong decisions. Thus, performance data of analytical methods are the main source of uncertainties. However, uncertainties obtained from limited observations should be used with caution to represent the overall uncertainty. This study demonstrated the influence of random uncertainties on the results of experimental approaches to estimate the uncertainty in chemical assays. The different profiles of uncertainty were demonstrated depending on the analytical technique and food composition.