News
New guidance on how to estimate the uncertainty that arises from sampling
By: Jessica Gowers
Last updated: Tuesday, 11 June 2019
An international team of scientists have published new guidance on how to estimate the uncertainty that arises from chemical sampling, allowing for greater confidence in decision-making.
The scientists led by Michael Ramsey, an Emeritus Professor from the University of Sussex, brought together organisations from across the world to produce the guidelines.
The Measurement uncertainty arising from sampling Guide describes various methods, some based on experimental results and others on theoretical modelling, which can be used to estimate the uncertainties arising from the process of sampling and the physical preparation of samples.
The Guide will allow specialists to make realistic evaluations of uncertainty, key for both scientific investigations and regulatory decision-making.
Michael Ramsey said: “It has been said that all measurements of chemical concentration are wrong, it’s just a question of by how much. This is because a measurement is only an estimate of the true value of concentration. It is crucial, therefore, to know the range within which the true value lies, which is called the measurement uncertainty.”
“One new idea from this second edition of the Guide is the ‘uncertainty factor’, which is a more accurate way to describe high levels of uncertainty. One of the examples in the Guide, on nitrate contamination of lettuce, shows that uncertainty estimated without considering the sampling was 8%, but was twice this value (16%) when sampling was included. Four batches of lettuce passed the test as uncontaminated when the sampling uncertainty was ignored, but three of these batches fail the test when the sampling uncertainty was included. This means that contaminated food could potentially be sold, if the uncertainty from sampling is ignored, or underestimated. Knowing the value of uncertainty can therefore lead to improved certainty in decision making.”
The Guide will also have consequences for risk assessments and describes a new approach that reduces the costs of estimating sampling uncertainty by a third.
It covers a wide range of different applications and includes a number of case studies. These include, but are not limited to, lead in contaminated land, levels of vitamin A in porridge, enzymes in chicken feed, levels of cadmium in agricultural soil and dissolved iron in groundwater.
The international organisations involved in the publication of the Guide are CITAC, Eurachem, Eurolab, Nordest and Royal Society of Chemistry’s Analytical Methods Committee.
Representatives of the 34 countries represented in Eurachem endorsed the Guidance in a recent vote.
There will be workshop to explain the importance of the Uncertainty from Sampling, the contents of the Guide, and examples of how it can usefully be applied, on 19-20th November 2019 in Berlin.
For more information, visit the Eurachem website.