G103 - A2LA Guide for Estimation of Uncertainty of Dimensional Calibration and Testing Results

Acknowledgements A2LA gratefully acknowledges the invaluable contributions of Mr. Ralph Veale to the Association over many years, both as an assessor and technical advisor, and for his work in writing this Guidance document for dimensional testing and calibration laboratories. We would also like to acknowledge the following individuals who have provided constructive input during the development of this document: Mr. Any errors or omissions remaining in this paper are, of course, the responsibility of the editor. Almost everyone makes measurements and almost everyone can ascribe some measure of uncertainty to this measurement. If we measure the area of a room by stepping off the length and width, we know that the number we get is not as " good " as if we had used a precision tape measure. We also know that if we are measuring the diameter of a crankshaft pin, we need a more accurate number than if we are measuring the diameter of a fence post to decide how big to dig the hole. Sometimes we even make uncertainty statements along with our measurements. If asked how good a particular measurement is, we may reply " right on " or some other colloquialism. If we know we just guessed, we may say " about six inches " or whatever the number was. We also know that sometimes it is economically feasible to use better equipment and be more accurate in our measurements. If we are only interested in the approximate amount of carpet we need for a room it is okay to step off the dimensions. But if we are buying expensive carpet we may want a more accurate number than we would get by stepping off the area. If we use a precision tape, take care in making the measurement, and check our arithmetic, we can be confident that we have a number good enough for our purpose. That is what uncertainty statements are about – a statement as to the confidence that we have in the measured value. The same reasoning applies for scientific measurements or for measurements made in a calibration or testing laboratory. National laboratories have for many years insisted that an uncertainty statement accompany all measurements. However, not all laboratories computed the uncertainty in the same manner nor reported the uncertainty with the same confidence level. For this reason a document (the " GUM ") was written to use as a …