ARCHIVED REPORTS_XR0008383 - Laserfiche WebLink

Home Browse Search

' Section 7 . ONSITE Environmental Laboratones Inc Revision 0 ' Quality Assurance Plan February 15, 1996 Page 3 of 4 7 2 1 Precision is the degree of reproducibility of a measurement and is evaluated by the analysts of samples and/or spiked samples in duplicate Precision is ' calculated as the "Relative Percent Difference", which is the range of difference between two measurements on the same sample divided by their mean, expressed as a percentage ' RPD = (Ongrnal Value - Duplicate Value) 100 1/ 2(Ortgtnal Value T Duplicate Value) For example, if the original result is 24 5 and the duplicate result is 24, then (24 5-24)/((24 5+24)/2) * 100 = 0 5/24 25 * 100 = 2 0 % RPD ' 7 2 2 Accuracy represents another measure of data quality It is the degree of agreement of a measured value of the analyte with its true or actual value It is evaluated by the use of samples to which known amounts of the target analytes have been added (called matrix spikes), known value control samples (blank spikes or Laboratory Control Samples), and standards derived from a source which is different from the calibration standards 7 2 2 1 In the case of second source standards, blank spikes, and control samples, the accuracy is calculated as a percentage of the true value, wherein a percentage over 100% indicates a high bias in the measurement and a percentage less than 100% indicates a low bias in the measurement, using the following equation I i Value Found ' %Recovery = 11 100 True Value For example, if an LCS True Value is 25 and the reading for the LCS was 24 5, then (24 5/25) * 100 = 98% Recovery 7 2 2 2 In the case of matrix spikes, the percent recovery is calculated using ' the following equation ' Spu(ce Recovery(/) = Observed r 100 Theoretical I Word 6 0 loc[=7 doc 1