Donald and Jan are right - SPM allows just 3 factors or less, plus a
replication factor. If you specify your design using "All scans" you
have to enter a nscan-by-4 index matrix. The replication factor goes
in column 1, and column 2-4 are the factors in the order specified in
your design. The "All scans" option is designed to bypass all SPM
auto-magic (e.g. for very special cases, where SPM might get confused
about the image<->factor relationships).
The limitation to 3 factors is really hardcoded in SPM, and this won't
change for SPM8.
Hope this helps,
Quoting "West, John D." <[log in to unmask]>:
> Thanks Donald.
> Ok, I can see how that would work, but doesn't this sort of limit
> you to just 3 factors or less? It doesn't seem like the flexible
> factorial method has that limitation.
> From: MCLAREN, Donald [mailto:[log in to unmask]]
> Sent: Friday, October 31, 2008 4:41 PM
> To: West, John D.; SPM
> Subject: Re: [SPM] Flexible factorial design error
> Since you have specified "All Scans", you need a factor matrix with
> 4 columns. There is no way for SPM to know which scans go with which
> subject if you leave it out. If you choose the other option, then
> you specify the other levels on a subject by subject basis and thus
> there is explicit knowledge of which scans go with which subjects.
> Column 1: should list the observation/row number
> Column 2: factor 1 level
> Column 3: factor 2 level (if no factor 2, use 1s)
> Column 4: factor 3 level (if no factor 3, use 1s)
> 1 1 1 1
> 2 1 1 2
> 3 1 1 3
> 4 2 1 1
> 5 2 1 2
> 16 6 2 1
> 17 6 2 2
> 18 6 2 3
> 27 9 3 3
> Hope this helps.
> On Fri, Oct 31, 2008 at 3:21 PM, West, John D.
> <[log in to unmask]<mailto:[log in to unmask]>> wrote:
> I am following Glascher and Gitelman's guide for flexible factorial
> design and have run into a snag.
> I am trying to set up a 3x3 factorial design where I have three
> groups and three time points. I want to include the main effects of
> subject, group, time and groupxtime interaction.
> So I set up factors for subject, group, time and choose to specify
> all scans and factor matrix.
> I have input the subjects list in order of group then time.
> Therefore my factor matrix should be (since the subject factor does
> not have to be specified as it is a special factor):
> Group 1:
> 1 1
> 1 2
> 1 3
> 1 1
> 1 2
> 1 3
> Group 2:
> 2 1
> 2 2
> 2 3
> 2 1
> 2 2
> 2 3
> Group 3:
> 3 1
> 3 2
> Then I set three main effects for subject, group, and time, and the
> interaction of group x time.
> This seems to be the right way to set things up but when I go to run
> this and create the design matrix I get the following:
> Error: factor matrix must have four columns.
> I am not sure why I get this error as I should not need four columns
> (I only have three factors overall and the subject factor should
> not need to be set). Can anybody tell me where I might be going
> Thanks in advance for the help!
> -John West
> IU School of Medicine
> Best Regards, Donald McLaren
> D.G. McLaren
> University of Wisconsin - Madison
> Neuroscience Training Program
> Office: (608) 265-9672
> Lab: (608) 256-1901 ext 12914
> Cell: (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.