Analysis Of Variance Interobserver Agreement

Kottner J, Audigé L, Brorson S, Donner A, Gajewski BJ, Hróbjartsson A, et al. Guidelines have been proposed for studies on the reliability of reports and agreement (GRRAS). J Clin Epidemiol. 2011;64 (1):96-106. Fortunately, the industrial era has given us a lot of experience in dealing with these problems by developing a process called Reproducibility Gauge and Reproducibility Evaluation, which has been relatively recently updated using ANOVA statistics (1). Eliasziw et al. (8) have transposed these methods into the field of medicine. In short, the method uses a two-way ANOVA to calculate intra-server and inter-observer SEM from a data set containing repeated measurements (tests) performed by several observers (evaluators). Like cci, SEM calculation can also be performed in cases that involve multiple measures and that observers are treated as both random and fixed effects.

. . .

Comments are closed.