Comparison of the performance of fourteen statistical tests for detection of outlying values in geochemical reference material databases

This work presents a comparison of relative efficiency of fourteen statisti cal tests to detect outliers in normally distributed samples. These tests i nclude deviation/spread, Grubbs-type, Dixon-type, and high-order moment sta tistics. Performance for the statistical tests is evaluated using Geochemic al Reference Material databases from the United States Geological Survey. T est efficiency is compared for the first application of statistics in versi on k = 1 and k = 2, as well as for tests in version k = 1 applied consecuti vely against block procedures in which k = 2, 3, or 4 values are evaluated at the same time. In both evaluations, the sensitivity of the statistical t ests shows a general strong dependence on sample size. The best performance is observed for the block procedures compared to consecutive statistical t ests affected by masking effects.