 |
 |
Since this appears to be a repeated measures design, then the subject factor is important to include. See https://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0803&L=SPM&P=R16629 for more details. The attachment in that archive also deals with the issue of variance. On Wed, Jan 21, 2009 at 2:54 PM, Cyril Pernet <[log in to unmask]> wrote: > Dear Roberto > > > I'm actually studying a group of patients with Parkinson disease > implanted > > with a cerebral stimulator (DBS) who perform a motor task. I'm wondering > if > > what I do is correct: this is a multisubject design and I think I should > > perform a full-factorial design on SPM5, using three factors: subjects, > > Stimulation (on vs. off) and Task (Movement vs. Rest). > > I see only 2 factors Stimulation and Task ; unless you also have controls > ?? > > > The questions are: > > 1) I read published papers in the literature with similar studies using > the > > default 'proportional scaling' while I read on the manual to use 'AnCova' > as > > normalization. Which is the correct one? > > depends what you want to do? if you have a measure, say of tremor, for each > subject with the stimulation on and off then you could use an AnCOVA to > regress this out ; this would give you the effects you are investigating > giving that the tremor differences between subject and condition (on off) is > taken into account .. there is no direct relation between proportional > scaling and AnCOVA .. > > > 2) When dealing with 'subject' as a new factor, do I have to use 'Equal' > > under 'Variance'? Most people dealing with SPM told me to leave > 'Unequal'. > > How may I manage this? Is 'Unequal' uncorrect? > > if it is for control vs patients then unequal, you don't expect the two > groups to be the same > > > > Why do you think there is this difference between theory and practice > > (published papers?) > > > ?? maybe people don't read the manual :-) as you did > > > > 3) One more question is: should I leave blank on 'Overall grand mean > > scaling' or select 'yes' at 50? > > as you want really, if you want to select a value take 100; the values in > the con images would then reflect the % of variation of the signal BOLD > > Good luck > > Cyril > > > -- > The University of Edinburgh is a charitable body, registered in > Scotland, with registration number SC005336. > -- Best Regards, Donald McLaren ===================== D.G. McLaren University of Wisconsin - Madison Neuroscience Training Program Office: (608) 265-9672 Lab: (608) 256-1901 ext 12914 ===================== 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 (608) 265-9672 or email.