Assessing Reliability Of Measurements With Generalizability Theory: An Application To Inter-Rater Reliability

Abstract

This article introduces the application of Generalizability Theory to assessing the reliability of measurements. Generalizability Theory can be used to assess reliability in the presence of multiple sources of error and also can be applied to assess reliability in the presence of different types of sources of error. In particular, the application of Generalizability Theory to measurements involving multiple raters is considered. An example illustrates its application to assessing reliability in the presence of inter-rater variability.