The chemistry laboratory at the Peruvian Institute of Nuclear Energy (IPEN) has carried out a validation method for the samples of siliceous composition. At least seven variables affecting the robustness of the results were initially identified, which may interact simultaneously or individually. Conventional evaluation hereof would imply a massive number of analyses and a far more effective approach for assessment of the robustness for these effects was found in the Youden-Steiner test, which provides the necessary information by only eight analyses for each sample type. Three reference materials were used for evaluating the effects of variations in sample mass, irradiation duration, standard mass, neutron flux, decay time, counting time and counting distance. © 2008 Springer Science+Business Media, LLC.