I estimate 600 samples will be need to be coded. The principal investigator will code all 600; however, how many of the samples will need to be coded by how many coders to minimize the variance of the intraclass correlation coefficient (ICC)?
Thank you for responding; however, I am receiving an error message and DOI not found when I click the link. I am very interested in reading the article. Could you please send the article title and authorship. Thank you so much!
After some detective work I believe this is the reference Ramtin is referring to;
S. D. Walter Dr., M. Eliasziw, A. Donner, Sample size and optimal designs for reliability studies, Statistics in Medicine, Volume 17, Issue 1, Pages 101–110, 15 January 1998