 |
 |
FYI, I addressed this question in a personal email. If others have questions, feel free to email me directly. Cheers, Jeanette On Fri, Feb 24, 2017 at 3:24 AM, Rosalia Dacosta Aguayo <[log in to unmask] > wrote: > Sorry for double posting....but I really need help with this issue... > > ---------- Forwarded message ---------- > From: Rosalia Dacosta Aguayo <[log in to unmask]> > Date: 2017-02-24 8:33 GMT+01:00 > Subject: Power Analysis Tool - Jeanette Mumford / Anderson Winkler > To: FSL - FMRIB's Software Library <[log in to unmask]> > > > Dear FSL experts and Anderson Winkler, > > I am planning to do a power analysis before carry on a project. I have > found the power analysis tool of Jeanette Munford: > https://www.nitrc.org/account/lostloginname.php (latest release). Now, > the problem is that I have to describe which kind of analysis I am going to > do in order to get the sample size needed to do this analysis and get the > power D and the p value associated. I have never done such analysis and I > am a bit confused about it. So I would be more than grateful if you could > enlight me with some explanations (I have a pdf with instructions as well). > > I have three experimental groups of randomized, parellel patients with > different treatments. For group one, treatment consists in A + B, for group > two, treatment consists in A + C, and for treatment three the treatment > consists in A + D. A + C is gong to be my control / reference group. I > would like to look for differences in MRI when adding B, C and D, if there > are interactions effects as well as additive effects. > > I need to translate this into a Model for the power analysis software and > I do not know how and how to make this software works....any explanation > about the model to follow and how to proceed with the software would be > greatly appreciate. I have a colleage that is enginyeer and he tried it but > he could not cope with it neither. > > Kind regards, > Rosalia > >