Calculation of Values - Essay Example

The following data are considered not valid for analysis: results that were reported erroneously (reporting errors) such as data with inappropriate units and "results reported as approximately 10, 100 or 1000 times greater or smaller than the majority of submitted results" (Croucher, 2004); results reported as not detected (n.d.); and results from participants not quoting a percentage recovery. Aside from these non-valid data, each report has the following points to be considered for removal of non-valid data.

Results for total fat recorded without any reference to the use of an acid hydrolysis step; and nitrogen results that appeared to be for protein i.e. results approximately a factor of 6 greater than the majority of results (Knaggs, 2004a). Results reported as 0 ug/kg; results from participants not quoting a percentage recovery or stating na, EXCEPT for total AF, where participants submitted results that were the sum of their recovery corrected individual aflatoxins (Walker, 2005). Participants reporting not detected and subsequently assigned a result of 0g/kg; results from participants whose recovery is outside the range 70-110%; and results below the stated LOQ including results with no LOQ stated (Leach, 2005).

The value of p estimates the limits . t of 0g/kg; results from participants whose recovery is outside the range 70-110%; and results below the stated LOQ including results with no LOQ stated (Leach, 2005).ii) Minimizing the influence of outliers by the use of a robust statistical procedure to derive the meaniii) Normality (Kolmogorov - Smirnov test), or otherwise, of the distribution of the results is considered.iv) Assessment of the standard uncertainty () of the robust mean = /n, = the standard deviation of the robust mean n = the number of data points used to calculate the robust meanB.

Target Standard Deviation for the RoundThe value of p estimates the limits of a satisfactory performance in a FAPAS proficiency test (FAPAS, 2002). In cases where the value of p does not reflect best practice for the analyses in question, the predictive models of the appropriate form of the Horwitz equation are valuable indicators of the best practice. This equation predicts a standard deviation from a given concentration (c) and requires c to be expressed as a dimensionless mass ratio (1ppm = 10-6 or % = 10-2)i) For analyte concentrations < 120 ppb = (0.22c) / mrii) For analyte concentrations 120 ppb and 13.

8% = (0.2c0.8495) / mr iii) For analyte concentrations > 13.8 % = (0.01c0.5) / mrwhere,c = concentration (assigned value), expressed as a dimensionless mass ratio mr = dimensionless mass ratio.In all reports, the was calculated according to the equations above. In report 0141 however only the value of for the ash was estimated in this manner. The value of for moisture, total fat, nitrogen and hydroxyproline was derived from collaborative trial data assuming a linear relationship : = (RSDR / 100)cwhere, RSDR = Relative