 |
 |
A couple of comments (3 being the most important): (1) Include all the main effects (group and task). (2) The contrast you stated seems to have an space or column orders reversed. In your design is subject first and the interaction second? What is the order of the interaction? Try: *Ones(1,16)/16 –ones(1,16)/16 ones(1,2)/2 **-ones(1,2)/2* *Ones(1,16)/16 –ones (1,16)/16 1/2 -1/2 1/2 -1/2* (3) Between-subject effects have the wrong statistics, they are INVALID. Anytime you have a between- and within-subject factor in the same model in SPM the statistics are inflated. This was the subject of my OHBM poster ( www.martinos.org/~mclaren/ftp/presentations/). In the poster, we concluded that you need to average over the conditions to get the proper between-subject effects or use our custom code (under development). We hope to have a beta version ready for testing shortly. Also, N-way within-subject effects are also INVALID (see post by Aaron Schultz). Hope this helps. Best Regards, Donald McLaren ================= D.G. McLaren, Ph.D. Postdoctoral Research Fellow, GRECC, Bedford VA Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School 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 Fri, Aug 12, 2011 at 9:01 AM, Per Nordmark <[log in to unmask]> wrote: > Dear SPM-Listers. > > I am trying to put together an ANOVA of 2 groups with 16 subjects in each, > I have got two factors (’Task’). For this I am following the “Contrast > weights in flexible factorial design with multiple groups of subjects” by > Jan Gläscher and Darren Gitelman and the section “Two groups of subjects”. > Everything works out fine until the modeling of the contrast weights for the > groups (“Design1”). For some reason the main effect of group does not work: > > For main effect I have chosen subject (factor 1). Interaction 2 3 (group x > ‘Task’). > > In the example in the tutorial there are different number of subjects in > the two groups (5 and 6) whereas in my data-set it is balanced (16 in each > group), givíng the contrast weights: > > *Ones(1,16)/16 –ones (1,16)/16 ones(1,2)/2 -ones(1,2)/2.* > > Modeling and performing the main effect of condition is no problem: > > *zeros(1, 16+16) 1 -1 1 -1* > Has anyone come across the same problem or am I doing something obviously > wrong? > > /Per >