Published: 01 January 1987
| ||Format||Pages||Price|| |
|PDF (92K)||6||$25||  ADD TO CART|
|Complete Source PDF (3.4M)||220||$55||  ADD TO CART|
Cite this document
During the past ten years, the importance of assessing the precision and the accuracy of data derived from standard methods has become recognized. Both the American Society for Testing and Materials (ASTM) and the U.S. Environmental Protection Agency (EPA) have developed policies requiring precision and accuracy assessment for methods before they are designated as standard methods or reference methods. The difficulty in implementing these policies is not in developing a meaningful uncertainty estimate for a given method, it is in separating the random from the systematic components of the error. In this paper, a method of separating error components into precision and bias is given. The method uses calibration data taken repeatedly at a fixed point using a standard whose total uncertainty is small compared to the uncertainty of the measurement method.
calibration, uncertainty estimate, precision, accuracy, data analysis, sampling, air quality, atmospheric measurements
Supervisory research chemist, National Measurement Laboratory, National Bureau of Standards, Center for Analytical Chemistry, Gas and Particulate Science Division, Gaithersburg, MD