Chapter 1: Preliminary Description of Error Analysis

No measurement can achieve absolute certainty, regardless of instrumental precision, due to both practical limitations in reading scales and fundamental problems of definition such as the non-uniformity of physical objects. The chapter emphasizes that measurements become meaningless without clearly stated margins of uncertainty, illustrated through the example of density measurements used to authenticate a gold crown, where one measurement's wide range proves inconclusive while another's narrow range definitively answers the question. Understanding and properly estimating uncertainty is critical across applied sciences, where engineers must account for material variability and human factors for safety, and in basic sciences, where experimental precision determines whether theories can be tested and compared, as demonstrated by the 1919 eclipse experiment that validated Einstein's general relativity. The chapter presents practical methods for estimating uncertainty: when reading marked scales, interpolation between markings with explicit uncertainty statements prevents false precision claims, while repeated measurements of the same quantity provide insight into uncertainty through observable data spread and averaging. A crucial distinction emerges between random uncertainties, which averaging can mitigate, and systematic errors, which consistently bias all measurements in one direction and cannot be detected through repetition alone but require calibration against more reliable standards. The author advocates for always explicitly stating uncertainties rather than relying on implied conventions, as proper uncertainty estimation forms the backbone of valid scientific conclusion and prevents dangerous practices such as uncritically accepting all significant figures from calculators.