Laserfiche WebLink
Project <br />Plan <br />Revision 6 <br />1/8/1988 <br />page 6 of 37 <br />special methods must be developed to measure an analyte in a particular <br />matrix, at acceptable precision, accuracy, and sample verification. <br />4.1.2.2 Method Verification <br />As a part of the procedure for method development, data obtained from <br />analysis of known standards solutions are examined for precision and <br />accuracy. Generally these data are generated by one or more analysts, each <br />measuring multiple replicates in the low and high ranges of the analysis in <br />order to provide statistically significant data. Percent recoveries and <br />standard deviations are calculated from these data for comparison with <br />those of previous methods. <br />To be accepted for use by the laboratory, the new method must first <br />demonstrate precision and accuracy which are acceptable according to <br />any <br />EPA guidelines available for the method. If such guidelines are not <br />available, the method must either show improved precision and accuracy over <br />previous methods or otherwise demonstrate superiority. <br />Before analytical data may be reported by the new method, it must also <br />be verified by performance samples. These may come from several sources <br />but it is important that the source be reliable and originate outside the <br />analytical group. Certification samples from Federal and State agencies <br />may be used if available. EPA Reference Solutions are commonly used and <br />i <br />are very reliable. Commercial Reference solutions may be used if EPA <br />solutions are not available. If the analyte is so uncom mon as to not be <br />found in any of these solutions, the sample may be prepared by the QA <br />officer. These samples should be made from standard stock solutions that <br />are separate from those used by the analysts. <br />12 <br />