 |
 |
Thank you so much for replying! I think that the suggestion of analyzing individuals' fmri data to find if any are wildly increasing the variance is excellent. By "eyeballing" I meant that when I look at the results of the Control A-B scans, and then I look at the Patient A-B scans, they look very different (with corresponding regions of activation that agree with our hypothesis). It looks very convincing, but I wanted to show that statistically by doing the 2-sample t-test. Thank you again! I really appreciate it. Omer Liran On Mon, Sep 8, 2014 at 1:12 PM, MCLAREN, Donald <[log in to unmask]> wrote: > The two sample t-test you are currently using is the correct way. If you > change to put each session into the group model, then you would violate the > assumption of independence and have inflated statistical results. > > How are you "eyeballing" the differences? > > I would also extract the data and see if there is one or two subjects that > are increasing the variance and causing the non-significant effects. > > Best Regards, Donald McLaren > ================= > D.G. McLaren, Ph.D. > Research Fellow, Department of Neurology, Massachusetts General Hospital > and > Harvard Medical School > Postdoctoral Research Fellow, GRECC, Bedford VA > Website: http://www.martinos.org/~mclaren > Office: (773) 406-2464 > ===================== > This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED > HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is > intended only for the use of the individual or entity named above. If the > reader of the e-mail is not the intended recipient or the employee or agent > responsible for delivering it to the intended recipient, you are hereby > notified that you are in possession of confidential and privileged > information. Any unauthorized use, disclosure, copying or the taking of any > action in reliance on the contents of this information is strictly > prohibited and may be unlawful. If you have received this e-mail > unintentionally, please immediately notify the sender via telephone at > (773) > 406-2464 or email. > > On Mon, Sep 8, 2014 at 4:00 PM, Omer Liran <[log in to unmask]> wrote: > >> Dear Experts, >> >> I want to make sure I am doing this right: I have 2 tasks (1 and 2) with >> 2 conditions in each (A and B), and two groups (control and patient, n=30 >> each). Each task is performed three times by each subject. For a first >> level analysis, for every task I add the 3 sessions and set up the >> conditions and contrasts to get A-B and B-A. I then want to compare the >> differences for control vs patient, so I perform a 2-sample t-test using >> the contrasts I created (n=30 each group). However, even though eyeballing >> the contrasts show clear group differences, the 2-sample t-test doesn't >> show significant differences. Would it be statistically wrong to analyze >> each session separately, and then drop them all into the 2-sample t-test >> (so I would have n=90 for each group since each session was repeated 3 >> times). Or is there a better way to do this? >> >> Thank you so much for your time! >> Omer Liran >> > >