Hello all. I'm working on a PhD thesis analyzing reproducibility/repeatability of a nuclear medicine examination. I have two sets of data (50 patients each) and 3 observers. In each case (patient), two variables are evaluated, first with a range of 0..100 and second 1..3. Each case is also evaluated twice by each observer (for additional intraobserver reproducibility). I'm using R package psy to calculate ICC (the other method I'm using is repeatability). My question with ICC is: how do I evaluate if the difference between observers (and patient groups) is statistically significant? I have been searching through the R archives, but I haven't been able to find an answer. Thank you, Luka Lezaic -- Nobody is perfect. My name is nobody. ------------------------------------------------------------- This mail was scanned by BitDefender For more informations please visit http://www.bitdefender.com