Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Stephen Henry <henrstep@med.umich.edu> |
To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
Subject | st: calculating class 3 ICC in Stata. |
Date | Wed, 9 Mar 2011 17:15:24 -0500 |
Hello, I have a question about calculating an ICC for raters in Stata. Specifically a "class 3" ICC as described by Shrout and Fleiss, 1979 in which raters are treated as fixed effects. In my data, the same 3 raters have each rated trust on different video clips. The data are in long form and each clip has 3 ratings by the same 3 raters. Most ICC formulas are calculated from mean-squares; I can get these using -anova trust clip rater- but Stata manual does not discuss how to specify fixed effects or crossed designs. Other ICC formulas are based on variance estimates, but I'm not sure how to get the appropriate variance components from -xtreg- or -xtmixed- Thanks Stephen Henry ********************************************************** Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/