Beyond the opposition between the Bayesian and the frequentistic views in measurement

The probabilistic modelling of measurement systems is discussed, showing how it provides a practical and effective way for addressing uncertainty evaluation, that does not require any commitment to specific philosophical schools, such as the frequentistic and the Bayesian ones. Firstly, different ways of modelling the same measuring device are compared and discussed, considering both the designer's and the user's standpoints. Then philosophical issues concerning both the nature of probability and the probabilistic inferences related to the measurement process are considered, in the perspective of this model-based approach. Applications to education and to the revision of international guides for measurement are finally addressed.