Hi Alisha,

Please, see below:


On 27 June 2014 17:30, Alisha Janssen <[log in to unmask]> wrote:
Hi Mark,

Sorry that I'm not Mark ;-)
 

I'm very sorry. I did a multi-part message so I could attach my design file, but wrote the actual message in the plain text box and it must not have sent. I'm running a repeated measures ANOVA (2x2 design). After looking through the message boards, I'm pretty clear on the GLM I need to set up, but would appreciate verification that this is the right design, and would like clarification on one point. My design is an intervention study with an experimental and control group. The task is an event-related design, and I have already run the first-level analysis (see the design file previously attached). The first-level produced three lower-level copes. Thus, I believe my higher level design would look similar to the example on the FEAT user page for repeated measures ANOVA, 2x2 design. For example, with two subjects (1 control, 1 experimental) my design file would look something like:


                                     Group      EV1(Time)         EV2(Group      EV3(Interaction)    
Input1(subject1-pre)           1                1                     1                        1                                             
Input2(subject1-post)         1               -1                     1                        -1                                            
Input3(subject2-pre)           1                1                    -1                        -1                                            
Input4(subject2-post)         1               -1                    -1                         1                                            

The contrasts would then be:

                                     EV1            EV2        EV3        
ME Time                         1                 0             0       
ME Group                        0                 1            0
GroupxTime                     0                 0            1


My questions are, is this the appropriate set-up for my analysis?

You need to add one extra EV per subject, as in the example: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#ANOVA:_2-groups.2C_2-levels_per_subject_.282-way_Mixed_Effect_ANOVA.29
 
And, I also noticed a fourth EV on the FEAT User guide example of 2x2 ANOVA with all 1's. I wasn't sure what the purpose of the 4th EV was and whether I would need it in my analysis. If you could provide further clarification on these points, I would greatly appreciate it. Thank you very much for your time.

In the 2x2 ANOVA (without repeated measurements), that EV codes for the grand mean. Here, however, with the repeated measurements, there isn't a grand mean, but instead the mean is coded by multiple EVs, one per subject. These are the EVs 3-6 in the example of the link above.

Hope this helps!

All the best,

Anderson



On Fri, Jun 27, 2014 at 3:33 AM, Mark Jenkinson <[log in to unmask]> wrote:
Hi,

I'm afraid that there was no question in this email, just an image.
Can you re-post the question?

All the best,
        Mark



On 25 Jun 2014, at 22:39, Alisha Janssen <[log in to unmask]> wrote:

> <firstlevel_post.png>



--
Alisha Janssen
Doctoral Student of Clinical Psychology
Clinical Neuroscience Laboratory
The Ohio State University
1835 Neil Avenue