03-04-2015 05:39 PM
I have a question about sample size calculation at the very beginning of study. Now the only thing I know is we will use three centers to collect data and use mixed model by the end. We are interested in patient satisfaction for visiting hospital when using a specific system. We are writing letter of intent now for fund application. I think I should set power as 80% and type I error as 0.05, but I don't know how to deal with the group difference and standard deviation. I think we will compare satisfaction level (percentage) between two groups. The other thing is about ICC, since there are 3 centers involved. The last thing is I am not sure how to do sample size calculation for TWO-LEVEL data. I think I should consider level II, as I know the number of sites, which is 3, and also there are not so many sites. Please give advice.
Thanks in advance!!