PESTICIDE ANALYSIS IN-GROUND WATER - STATISTICAL EVALUATION OF CERTIFICATION DATA OF A MULTICOMPONENT REFERENCE MATERIAL

21 laboratories participated in the certification of a multicomponent ampouled reference material containing 24 pesticides. This paper descr ibes the statistical analysis of the unbalanced data set returned by t he participants in the certification. The simplest method of certifica tion is achieved when laboratories report all the required data. In pr actice, however, with complex analytical methods such as multicomponen t analyses of organic compounds, laboratories are not able to report a ll the required data. The response for each determinand in the materia l has been modelled in a three-factor nested model, For computational comfort the model has been rephrased. into a general linear model. The statistical problems with unbalanced data are discussed in relation t o ISO 5725 and ISO Guide 35. Tests for homogeneity are exact and uninf luenced by the unbalance in data. Unbalanced data cause dependencies b etween mean squares above the second lowest level of the model, thus e xact hypothesis testing is not possible at higher levels. Approximate tests for variation between rounds and laboratories has been construct ed by means of Satterthwaite's formula [I]. A graphical plot of devian ces from certified values has been applied to illustrate laboratory pe rformance.