Received: 28 May 2003 Accepted: 4 June 2003 Published online: 7 August 2003 © Springer-Verlag 2003 H. Fostel BMwA—Abt. IX/2, Landstraßer Hauptstr. 55–57, 1030 Wien, Austria I. Papadakis Department of Chemistry, University of Cyprus, P.O.Box 20537, 1678 Nicosia, Cyprus M. G. del Monte EURACHEM Italy, Viale Aeronautica 99, 00144 Roma, Italy N. Boley LGC, Queens Road, Teddington, Middlesex, TW11 0LY, United Kingdom P. M. van Berkel Kiwa Water Research, P.O. Box 1072, 3430 BB Nieuwegein, The Netherlands pects that should be taken into ac- count when trying to make conclu- sions from the relative performance of accredited laboratories in profi- ciency testing. Keywords Proficiency test exercises · Accredited laboratories · Relative performance Accred Qual Assur (2003) 8:511–513 DOI 10.1007/s00769-003-0674-0 GENERAL PAPER L. Cortez A. Duarte A. Hundewadt A. Schmidt B. Steffen D. Tholen H. Fostel I. Papadakis M. G. del Monte N. Boley P. M. van Berkel How to interpret information from proficiency test exercises concerning the relative performance of accredited laboratories Introduction Laboratories have to produce results precise and accurate enough to be used by the customer for the intended pur- pose. However, even experienced laboratories may pro- duce occasionally wrong results. This can happen due to several factors, for example human mistakes, causing random and systematic errors. Accredited laboratories also may produce wrong re- sults. Nevertheless, whilst accreditation cannot guarantee the quality of all the data produced by the laboratory, it should decrease the frequency of wrong results. A complement to accreditation is given by proficien- cy testing (PT), which allows a laboratory to compare its results with those of other laboratories on samples which should be similar to those routinely analysed by the labo- ratory using its routine methods and procedures [1, 2] Participation in PT is therefore necessary, especially when other quality control tools and proper traceability chains are not available [3]. L. Cortez ( ✉ ) EEE-PT Chairman, IPQ, Rua Antonio Giao 2, 2829–513 Caparica, Portugal e-mail: lcortez@mail.ipq.pt Tel.: +351-21-2948100 Fax: +351-21-2948188 A. Duarte EEE-PT Secretariat, RELACRE, Rua Filipe Folque 20–6ºD, 1050–113 Lisboa, Portugal A. Hundewadt DANAK, Tagensvey 137, 2200 Kopenhagen, Denmark A. Schmidt EUROLAB/BAM, Unter den Eichen 87, 12205 Berlin, Germany B. Steffen DAR/BAM, Unter den Eichen 87, 12205 Berlin, Germany D. Tholen Statistical Consulting Services, 823 Webster Street, Traverse City, Michigan, 49686, USA Abstract If the results from various proficiency tests are used to compare the performance of accredited and non-accredited laboratories, great care must be taken not to come to contradicting conclusions when in- terpreting them. This paper intends to draw the attention to various as-