 |
 |
Conducting a post-hoc t-test to explain a significant F-test is not double dipping. It is simply an extension of the same analysis. However, you want to be clear that you are only testing the areas that had a significant group effect and only to show which groups are different. With the F-test, you already know that there is at least one group that differs from the others. The issue of double-dipping arises, when you select your cluster based a correlation, then extract the ROI to do the same correlation. This is double-dipping as you've done the analysis and are doing the same analysis on a subset of the data. The way around this issue is to plot the ROI, but make the inference from the original analysis. 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 Thu, May 30, 2013 at 1:40 AM, Ce Mo <[log in to unmask]> wrote: > Dear all, > > I am wondering whether I am in the danger of circularity in my ROI > definition process and I'd appreciate any statistical insights. > > In my study I have 3 conditions of interest. As a first step, I conducted > an one-way within-subject ANOVA analysis and performed an F contrast to > test the condition effects (i.e. main effect), which gave me several > functional clusters. However, I am not sure whether I could conduct > subsequent ROI analysis on these clusters due to the worry of possible > danger of "double dipping" in my ROI definition. > > If I understand the concept of 'Circularity'/'Double Dipping' correctly, I > should avoid performing the same analysis again on the data > "selected" based on the results of this analysis that has been conducted > prior to it. However, I wonder whether F test and T test implemented in SPM > are different analytical approaches, in which case I assume I am good ? > > Many thanks and best regards, > Ce > > >