Hi Doug,
Thanks for your quick response - that all makes sense now.
Best wishes,
Clare
------------------------------
Message: 13
Date: Fri, 10 Aug 2012 10:19:43 +0000
From: "Gibbard, Clare" <c.gibbard(a)ucl.ac.uk>
Subject: [Freesurfer] QDEC results
To: "freesurfer(a)nmr.mgh.harvard.edu" <freesurfer(a)nmr.mgh.harvard.edu>
Message-ID:
<D2ACCCB2F2AEDD40865D0836BF8D33B6224C82C9(a)DB3PRD0104MB131.eurprd01.prod.exchangelabs.com>
Content-Type: text/plain; charset="iso-8859-1"
Hello,
I went on the FS course in Amsterdam recently. I have run a QDEC analysis to see whether cortical thickness differs between my two groups. I have two questions:
1. I put in the lh_bankssts_thickness as my continuous variable. The display outputs results for whether average thickness differs between my groups. The significant clusters appear in areas which are far away from the lh_bankssts. Is this correct? Does the design somehow look at overall cortical thickness in the lh even though I have only specified the lh_bankssts?
2. I can't see the button for controlling for multiple comparisons. From the wiki I believe this is supposed to appear under a scalar heading, but I can't see it. Do I need to add something in order to do this?
Thank you very much.
Best wishes,
Clare
----------------
Message: 21
Date: Fri, 10 Aug 2012 11:03:52 -0400
From: Douglas N Greve <greve(a)nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] QDEC results
To: freesurfer(a)nmr.mgh.harvard.edu
Message-ID: <502522D8.2080307(a)nmr.mgh.harvard.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi Clare,
On 08/10/2012 06:19 AM, Gibbard, Clare wrote:
> Hello,
>
> I went on the FS course in Amsterdam recently. I have run a QDEC
> analysis to see whether cortical thickness differs between my two
> groups. I have two questions:
>
> 1. I put in the lh_bankssts_thickness as my continuous variable. The
> display outputs results for whether average thickness differs between
> my groups. The significant clusters appear in areas which are far
> away from the lh_bankssts. Is this correct? Does the design somehow
> look at overall cortical thickness in the lh even though I have only
> specified the lh_bankssts?
I think it is correct. You are not confining the analysis to bankssts.
You are basically using bankssts as a nuisance variable.
>
> 2. I can't see the button for controlling for multiple comparisons.
> From the wiki I believe this is supposed to appear under a scalar
> heading, but I can't see it. Do I need to add something in order to
> do this?
You should not need anything else. Are you on the "Display" tab? It
should be under the "Clusters" area.
doug