Some statistical procedures for analytical method accuracy tests and estimation.
Fischbach-T; Song-R; Shulman-S
Am Ind Hyg Assoc J 1996 May; 57(5):440-451
The accuracy criterion (AC) and statistical protocol developed by NIOSH for use with NIOSH and OSHA exposure standards were reviewed. A newly developed accuracy analysis approach was assessed and information of use in the development of further efficient, effective, and flexible method accuracy evaluations was provided. The NIOSH statistical model for accuracy and the associated AC were presented. The newly developed accuracy analysis model focused on the case of known bias and attempts to use this to estimate cases with unknown bias. Statistical procedures for addressing conditions of known and unknown bias and for accepting or rejecting the associated methods were also presented. The methods used in calculating critical coefficients of variation and estimating method inaccuracy, and for calculating confidence statistics for method inaccuracy were described. The investigation emphasized the use of general, flexible approaches in the design and analysis of reliable method accuracy models.
NIOSH-Author; Exposure-limits; Analytical-models; Mathematical-models; Statistical-analysis; Analytical-methods; Industrial-hygiene
American Industrial Hygiene Association Journal