Hi Monica,
Answers inline with questions below:
> On 30 Aug 2022, at 21:03, Monica Keith <[log in to unmask]> wrote:
>
> Hi, I see in the FslWiki that there is a model for "Two group difference adjusted for covariate" (option1) and "two groups with continuous covariate interaction" (option2). I have a couple of questions regarding these two models.
>
> 1. My understanding is that the first one is used to test if FA (my dependent variable) is different between the two groups controlling for some covariate (i.e. age). The second one is used to test if the differences in FA between group1 and group2 vary with age. Is this correct
Correct ( although there are a number of ways to describe the interaction effect, e.g. “the relationship between FA and age varies with group” )
> 2. The documentation says "if the interaction is significant, only the inferences of the interaction are of interest". Does this mean that if the second model gives significant voxels, I cannot test if there's differences between groups? So, I will never be able to say "FA is lower in patients than controls" because that difference is affected/depends on age?
Yes - inference on the group effect, if an interaction is present, is only valid for subjects of average age ( assuming the age covariate was demeaned before splitting ).
>
> Thanks
Kind Regards,
Matthew
--------------------------------
Dr Matthew Webster
FMRIB Centre
John Radcliffe Hospital
University of Oxford
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|