Marcus wrote in USMA 14418: >On Sun, 15 Jul 2001 20:16:18 > Bill Potts wrote: >>Marcus Berger wrote: >>> The zero correctly identifies the accuracy of the value that is >>> supposed to be reported. >> >>Precision, not accuracy. If that person is really, for example, 1.92 m tall, >>then 1.70 is hardly accurate. >>... >I'm sorry, Bill, but I don't think so! I do understand what you mean though. > >In our course at the U of A (University of Alberta) dealing with this >subject of measurements in the third year of engineering, that's the >technical term that is used all the time to mean the decimal place with >which instruments would measure certain physical properties. There is >even a diagram that is used to illustrate the concept in which "precision" >is associated with what you mean here, indeed, but not with the number of >decimal places a measurement is reported as. I could try to fetch that >material for you later on (once I have the chance to go there sometime in >the next few days). > >Marcus The Canadian Metric Practice Guide puts it this way: 10,2,6 Experimental Accuracy It is also important to distinguish between precision and accuracy. The accuracy of the measured value of a physical quantity is the difference the measured value and the true value. The accuracy is limited by systematic arrors in the experimantal method, while the precision is limited by the random errors in the experimental conditions. A highly precise measured value can be highly inaccurate if the measuring instrument is not properly calibrated. One needs to understand the technique sufficiently well to eliminate, correct, or allow for all possible sources of systematic error in any meqasurement. The reader will assume that the residual systematic error is less than the implied precision of the quoted result. It is the responsibility of the person performing the measrument to ensure that this is so. Joseph B. Reid 17 Glebe Road West Toronto M5P 1C8 Tel. 416 486-6071
