On the degree of objectivity of uncertainty evaluation in metrology and testing

Measurement uncertainty of transmission line resistance calculation using ‘Guide to the Expression of Uncertainty in Measurement’ and adaptive Monte–Carlo method

“…In the scientific literature, there appear a number of articles that deal with deficiencies and possibilities of improving the measurement uncertainty calculation procedure [11, 22, 23] recommended in the GUM [5]. According to Vilbaste et al .…”

The probability in throwing dices and in measurement

“…The accent on the subjective side of the uncertainty analysis, (see also [25]) may seem to rule out 'frequentist' methodology; however, this does not necessarily mean that the Bayesian one can always be better used instead. In Chapter 1 of [22], pages 30-31, the authors say: "Bayes theorem is an uncontroversial part of probability theory.…”

Data inter-comparisons in the context of the knowledge-gaining process: an overview

“…In certain situations or fields (especially in testing), comparisons may be made complicated either by the fact that no reference exists or because no person or organization is taking the initiative of setting up a comparison campaign. In that case, e.g., Shirono et al [17], Thompon et al [11] and Ponomareva et al [19] provide performance evaluation methods for PTs with uncertainty information when there is no reference laboratory available. In the absence of knowledge of a "true value", the fundamental aim of comparison measurements is to increase as much as possible the confidence level associated with the evaluation of the measurement result.…”

“…Note that the attribute "best" often associated to the evaluation of a measurement can only have the meaning associated with (or derived from) the hypotheses under which the analysis is performed, which vary according to subjective preferences. Without entering into the vast subject of subjectivity vs. objectivity (see, e.g., [19]), it is enough to note here that various degrees of subjectivity are possible, depending on the overall 'level of knowledge'-here a term not necessarily used in the Bayesian sense-associated with a specific measurand. Strangely, however, this fact is barely referred to, if not totally ignored, in the main approaches to statistics of interest to metrology and testing, i.e., the "error approach" and the more recent "uncertainty approach".…”