Comparaison de différentes stratégies d’évaluation de la performance dans les essais d’aptitude par comparaisons interlaboratoires
|Author(s)||Desenfant Michèle1, Allard Alexandre1, Lalère Béatrice1, Lardy-Fontan Sophie1, Le Diouron Véronique1, Lepot Bénédicte2, Ziegler Eric3|
|Affiliation(s)||1 : LNE, Laboratoire national de métrologie et d’essais, 1 rue Gaston Boissier, 75724 Paris cedex 15, FRANCE
2 : INERIS, Parc technologique ALATA, BP2 60550 Verneuil-en-Halatte, FRANCE
3 : Bipea, 189 rue d’Aubervilliers, 75018 Paris, FRANCE
|Meeting||17th International Congress of Metrology|
|Source||17th International Congress of Metrology, 02003 (2015). Section Maitrise de l’incertitude / Uncertainty management. 6p.|
Proficiency Testing (PT) uses interlaboratory comparisons to evaluate the performance of participants for tests or measurements. The statistical design and analytical techniques applied must be appropriate for the stated purpose of each PT. Most PT schemes compare the participant’s deviation from an assigned value with a numerical criterion. Different scoring strategies are available and commonly used, as described in the International Standard ISO 13528, with a new version scheduled for publication at the end of 2015. The different strategies used to determine the assigned value and the criterion to calculate performance statistics are critical because they will lead to different interpretations. It is very important to consider whether the assigned value and its standard uncertainty as well as the criterion are independent of participant results, or are derived from the submitted results. The two strategies will be presented, compared and discussed in perspectives of the concept of comparability and the concept of compatibility that are defined in the International Vocabulary of Metrology (VIM). To do so, a certified reference material was deployed in two proficiency testing schemes. They aimed at determining a set of relevant pesticidesbelonging to two chemical families: triazines and phenylurea - that are part of the list of priority substances of the European Water Framework Directive. Various methods to determine the assigned value from the results of participants were used : the algorithm A on one hand and the Q method and Hampel estimates on the other hand, as introduced in the new version of ISO 13528. Moreover, on the basis of the available estimators, the complementarity use of En and Z score in the perspectives of evaluating the performance of laboratories was addressed.