|Universiteit Faculteit FNWI English version||2 maart 2004, e-mail|
Woensdag 3 maart 2004
AbstractPart of the variation in observations arises within the measurement process.
Large measurement variation may lead to incorrect decisions, imprecise estimates or induce a large number of observations to diminish the risk of incorrect decisions or to increase the precision of the estimates.
The statistical evaluation of the measurement process is focused on the estimate of the variation due to the measurement process. The standard approach is to use analysis of variance models with random effects for all sources of variation.
However, often the statistical assumptions do not hold and adaptation to the analysis of variance models or alternative models are required. Some results on these alternatives will be discussed.
Furthermore, other results and ideas on the estimate of measurement variation, which are only in a preliminary phase, will be presented briefly.