Task Views
Measurement Uncertainty
Background
Measurement uncertainty is an important topic for all measurement fields, and analytical measurement is no exception. No measurement result can be interpreted correctly without at least some knowledge of the associated uncertainty; either the user needs to know how large the uncertainty is so that they can make appropriate allowances, or they need assurance that the associated uncertainty is sufficiently small to neglect for their particular purpose. Establishing sufficient knowledge and control of measurement uncertainty and communicating that knowledge to the client when required are therefore among the responsibilities of every laboratory.
Laboratories accredited to ISO/IEC 17025 or related standards such as ISO 15189 will be familiar with this responsibility, as these standards require laboratories to identify the main contributions to uncertainty and to provide estimates of uncertainty where appropriate. For calibration laboratories and reference material producers, the requirements are more stringent; uncertainties need to be evaluated following current international guidelines and reported on calibration certificates.
Resources
Eurachem provides a range of resources to assist laboratories responsible for evaluating, using or reporting measurement uncertainty associated with analytical measurement results:
- The Eurachem Measurement Uncertainty and Traceability working group develops guidance and coordinates workshops on measurement uncertainty for analytical measurement.
- The Eurachem guide "Quantifying uncertainty in analytical measurement (QUAM:2012)", currently in its third edition, gives detailed guidance for the evaluation and expression of uncertainty in quantitative chemical analysis, based on the approach taken in the ISO “Guide to the Expression of Uncertainty in Measurement”. The guide provides a range of approaches to uncertainty evaluation, including the use of method validation data as well as the detailed measurement model approach described in the ISO Guide. QUAM:2012 also gives details of essential statistical procedures (for example, the assessment of uncertainty from linear calibration), provides practical guidance such as cause-and effect analysis for identifying sources of uncerainty and a detailed description of uncertainty evaluation using spreadsheets, and includes detailed worked examples illustrating many common operations in uncertainty evaluation for chemical measurement.
- The Eurachem guide "Setting and Using Target Uncertainty in Chemical Measurement" discusses how to set a maximum admissible uncertainty, defined in the third edition of the International Vocabulary of Metrology as the “target uncertainty”, to check whether measurement quality quantified by the measurement uncertainty is fit for the intended purpose.
- A further guide, "Use of uncertainty information in compliance assessment", gives additional detail on the use of uncertainty information when assessing compliance with specifications, including legal limits.
- The Eurachem working group on uncertainty arising from sampling has published a first edition of the guide "Measurement uncertainty arising from sampling: A guide to methods and approaches". This guide describes different methods for assessing uncertainties related to sampling, including a simple and general method based on replicated sampling.
- Four information leaflets provide summary information on measurement uncertainty:
- The "Information leaflet for lab customers concerning the quality of chemical analyses", developed by SP, provides a simple explanation of measurement uncertainty for laboratory customers.
- The principles of the compliance assessment guide above are summarised in the leaflet "Use of uncertainty information in compliance assessment".
- The sometimes difficult question of when (and when not) to divide a standard deviation by the square root of a number of observations, in order to obtain a standard uncertainty, is covered in the leaflet "Using repeated measurements to improve the standard uncertainty".
- Another often difficult question, whether or not an observed significant bias should be corrected and the impact this may have on the measurement uncertainty, is discussed in a further leaflet on "Treatment of an observed bias".
- Eurachem has organised a number of workshops on measurement uncertainty and related topics, and the presentations are available for recent workshops. A list of workshops on this site can be found in the Completed Events list.
- Details
- Category: Task Views
- Last Updated: Tuesday, 03 March 2020 23:16