Some comments on the repeatability of measurements

Repeatability is the consistency with which an observer measures a set of objects. This statistic can therefore be used both to assess observer performance and to detect changes in the size of the objects between measurements. These uses are described in the context of biometric studies of birds caught by ringers. A simple account is given of how to calculate repeatability and to report the results. Several problems are discussed, with particular emphasis on those caused by the relationship between repeatability and the variability between specimens measured.