Statistica Evaluation of Some Analytical Methods in the Hydrogeochemistry Laboratory of General Directorate of Mineral Research and Exploration

 Several analytical methods are being employed related to various research projects and routine works on water samples in hydrogeochemistry laboratory of General Directorate of Mineral Research and Exploration. Some statistical evaluations are necessary in order to obtain accurate results for the projects related. In order to control the precision and accuracy of the chemical determinations in our laboratory, standard water samples have been requested from EPA (Environmental Protection Agency, USA). The samples donated by EPA have been analyzed for 28 species using our standard methods. Results of the determinations were compared with the true values supplied by EPA using «Studend's t test». R values have been used to investigate the presence of systematic errors at 95 % an 99 % confidence levels. Accuracy and precision of the analytical methods used have been statistically determined. No systematic error was observed for 43 results of the total of 54. Evaluation for systematic error could not be realized for 11 results. Relative percent errors were found to be in the ranges at 0-10, 11-20, 21-50 and more than 50 for 33, 10, 6,5 results respectively. As a result of this study, by applying the statistical tests for accuracy and precision, it has been determined that accurate results are obtained in most of the methods, only two methods require further development. 

___

  • Laitinen, H.A., 1960, Chemical Analysis: Mc. Graw Hill Book Co., Toronto 611. Langley, R., 1971, Practical Statistics: Dover Publications Inc., USA 399.