A simple method for evaluating data from an interlaboratory study

Large-scale laboratory- and method-performance studies involving more than about 30 laboratories may be evaluated by calculating the HORRAT ratio for each test sample (HORRAT = [experimentally found among-laboratories relativ e standard deviation] divided by [relative standard deviation calculated fr om the Horwitz formula]). The chemical analytical method is deemed acceptab le per se if HORRAT approximate to 1.0 (+/- 0.5). If HORRAT is greater than or similar to 2.0, the most extreme values are removed successively until an "acceptable" ratio is obtained. The laboratories responsible for the ext reme values that are removed should examine their technique and procedures. If greater than or similar to 15% of the values have to be removed, the in structions and the methods should be examined. This suggested computation p rocedure is simple and does not require statistical outlier tables. Propose d action limits may be adjusted according to experience. Data supporting U. S. Environmental Protection Agency method 245.1 for mercury in waters (manu al cold-vapor atomic absorption spectrometry), supplemented by subsequent l aboratory-performance data, were reexamined in this manner. Method-performa nce parameters (means and among-laboratories relative standard deviations) were comparable with results from the original statistical analysis that us ed a robust biweight procedure for outlier removal. The precision of the cu rrent controlled performance is better by a factor of 4 than that of estima tes resulting from the original method-performance study, at the expense of rejecting more experimental values as outliers.