CAC/GL 59-2006 Guidelines on Estimation of Uncertainty of Results
1. INTRODUCTION
It is a requirement under ISO/IEC 17025 that laboratories determine and make available the uncertainty associated with analytical results. To this end, food testing laboratories operating under Guidelines on Good Laboratory Practice in Pesticide Residue Analysis (CAC/GL 40-1993) should have available sufficient data derived from method validation/verification, inter-laboratory studies and in-house quality control activities, which can be applied to estimate the uncertainties particularly for the routine methods undertaken in the laboratory. These guidelines were prepared taking into account the general recommendations of the Codex Committee on Methods of Analysis and Sampling (CCMAS).
1.1 CONCEPT AND COMPONENTS OF UNCERTAINTY
Measurement uncertainty refers to the ‘uncertainty’ associated with data generated by a measurement process. In analytical chemistry, it generally defines the uncertainty associated with the laboratory process but may also include an uncertainty component associated with sampling.
The uncertainty ‘estimate’ therefore describes the range around a reported or experimental result within which the true value can be expected to lie within a defined level of probability. This is a different concept to measurement error which can be defined as the difference between an individual result and the true value.
The reporting of uncertainty is intended to provide a higher level of confidence in the validity of the reported result.
Contributions to data uncertainty are manifold and described in detail in Tables 1and 2. The evaluation of uncertainty ideally requires an understanding and estimation of the contributions to the uncertainty of each of the activities involved in the measurement process.
