| Zugriffsnummer | 29101 |
| Dokumenttyp | Konferenzartikel |
| Sprache | Englisch |
| Titel | Random effects ANOVA in uncertainty evaluation |
| Autor(in); Institution |
Arendacká, Barbora; 8.4, Mathematische Modellierung und Datenanalyse, PTB-Berlin
Täubner, Angelika; 1.7, Akustik und Dynamik, PTB-Braunschweig
Eichstädt, Sascha; 8.4, Mathematische Modellierung und Datenanalyse, PTB-Berlin
Bruns, Thomas; 1.7, Akustik und Dynamik, PTB-Braunschweig
Elster, Clemens; 8.4, Mathematische Modellierung und Datenanalyse, PTB-Berlin
|
| Quelle/Jahr | Measurement 2013, Proceedings of the 9th International Conference on Measurement:(2013), 5 S. |
| Herausgeber(in) |
Manka, J.
|
| ISBN | 978-80-969672-5-4 |
| Verlag | Bratislava: Institute of Measurement Science |
| Konferenzangaben | 9th International Conference on Measurement 2013, Smolenice, 27-30, May, 2013, Slovakia |
| Freie Schlagworte | random effects ; ANOVA ; type A uncertainty |
| Zusammenfassung | Random effects ANOVA models are well established and implemented in statistical software, they provide enough freedom to model dependencies between repeated measurements (e.g. a stronger dependency between measurements obtained within one day as compared to measurements obtained on different days) and they offer a closer look at the structure of the uncertainty (by splitting the variability due to different sources). The paper illustrates these concepts and advocates the use of random effects ANOVA models for analysis of long-term repeated experiments conducted to assess reproducibility of a measurement. Variability revealed by such an experiment is important when creating an uncertainty budget. The concepts are illustrated with the help of data coming from calibration of accelerometers. |
