| Zugriffsnummer | 25490 |
| Dokumenttyp | Konferenzartikel |
| Sprache | Englisch |
| Titel | Metrological compatibility and statistical consistency |
| Autor(in); Institution |
Knacker, Raghu; National Institute of Standards and Technology (NIST), Gaithersburg, MD, USA
Kessel, Rüdiger; National Institute of Standards and Technology (NIST), Gaithersburg, MD, USA
Sommer, Klaus-Dieter; 3, Chemische Physik und Explosionsschutz, PTB-Braunschweig
|
| Quelle/Jahr | Proceedings of the 10th ISMQC, the 10th International Symposium on Measurement and Quality Control:(2010), F2-157-1 - F2-157-4 |
| Availability | [CD-ROM] paper no.157 |
| Herausgeber(in) |
Takaya, Y.
|
| ISBN | 978-4-9905119-0-6 |
| Verlag | Tokyo: JSPE Techn. Committee for Intelligent Nano-Measure |
| Konferenzangaben | 10th International Symposium on Measurement and Quality Control (ISMQC-2010), Osaka, 05-09, September, 2010, Japan |
| Freie Schlagworte | Birge test ; Interlaboratory evaluation ; Uncertainty |
| Zusammenfassung | The Birge chi-square test of statistical consistency is the most widely used method to assess the significance of the differences between two or more uncorrelated measured values fo a common measurand. The International Vocabulary of Metrology, third edition (VIM3) introduced the concept of metrological compatibility which can be used to assess the significance of the differences between two or more results determined according th the Guide to the Expression of Uncertainty in Measurement (GUM). We discuss and compare the concept of statistical consistency motivated by the Birge test and the VIM3 concept of metrological compatibility. To assess the statistical consistency of a set of measured values a common practice is to pretend that the standard uncertainties are the known standard deviations of the presumed normal sampling distributions of the measured values from random variation in contemplated replications. This is inappropriate use of the standard uncertainties determined according to the GUM. In particular, the concept of statistical consistency is inappropriate for the results determined according to the GUM. Results of measurement which are not traceable to the same reference are not comparable metrologically. The concept of statistical consistency does not require that the measured values be traceable to the same reference. The conclusion of a test of metrological compatibility of two results does not conflict with the conclusion of a test of their statistical consistency. Thus there is every reason to use the concept of metrological compatibility in place of statistical consistency in the field of metrology. |
