See in line responses below.
On Sat, Dec 29, 2012 at 12:01 PM, Jinhui Wang
<[log in to unmask]> wrote:
> Dear SPMers,
>
> I have some questions on flexible factorial design in SPM8 as described in
> the following:
>
> Image a 2*3 factorial design with 2 within-subject factors A (2 levels) and
> B (3 levels) in 5 subjects, the factors are configured in flexible factorial
> design/SPM8 as:
> Factor 1: subject
> Factor 2: A
> Factor 3: B
>
> The ordering of each subject's images is: A1B1 A1B2 A1B3 A2B1 A2B2 A2B3,
> therefore, the conditions for each subject is set as: 1 1; 1 2; 1 3; 2 1; 2
> 2; 2 3 (according to "Contrast weights in flexible factorial design with
> multiple groups of subjects by Jan Glascher and Darren Gitelman).
>
> After running SPM, i was showed that the parameter estimability is not
> uniquely specified (regardless of the combination of effects to be tested),
>>> Repeated measure designs will never have uniquely estimable parameters because each condition estimate (e.g. A1B1) is dependent on the other conditions since you have a subject factor (which is necessary). This is exactly as expected and all it means is that there are multiple solutions to the equation. The important aspect though is that the relationship between conditions would not change with each possible solution.
> then i check the resultant factor matrix sorted in SPM.xX.I:
> 1 1 1 1
> 1 1 1 2
> 1 1 1 3
> 1 1 2 1
> 1 1 2 2
> 1 1 2 3
> 1 2 1 1
> 1 2 1 2
> 1 2 1 3
> 1 2 2 1
> 1 2 2 2
> 1 2 2 3
> 1 3 1 1
> 1 3 1 2
> 1 3 1 3
> 1 3 2 1
> 1 3 2 2
> 1 3 2 3
> 1 4 1 1
> 1 4 1 2
> 1 4 1 3
> 1 4 2 1
> 1 4 2 2
> 1 4 2 3
> 1 5 1 1
> 1 5 1 2
> 1 5 1 3
> 1 5 2 1
> 1 5 2 2
> 1 5 2 3
>
> In the SPM manual, it says "for eg. a two-factor design the first column
> denotes the replication number and columns two and three have entries like 2
> 3 denoting the 2nd level of the first factor and 3rd level of the second
> factor. The 4th column in I would contain all 1s".
>
> So, it looks that my factor matrix does not match the above statement: for
> my example, the first column contains all 1s.
>>> Your design matrix is fine. You have subject, factor1, and factor 2. In the manual, there are only 2 factors and you have three factors as subject counts as a factor.
>
>
> BTW, what does replication number exactly mean? In my opinion, it is related
> to the subject index, right?
>> It means the replication of the combination of three factors. Since each combination you have is unique, the replication number is always 1.
>
> Any comments are greatly appreciated!
>> You might want to check out GLM Flex (just google GLM Flex Aaron) for multiple with-subject factors.
>
>
>
> -- Jinhui WANG, Ph.D., Investigator
>
> Center for Cognition and Brain Disorders
> Hangzhou Normal University
>
> Add: No.276 Yuanli Building, Lishui Road,
> Hangzhou, 310000, PR China
>
> E-mail: [log in to unmask]
> [log in to unmask]
>
> Google Scholar Citations:
> http://scholar.google.com/citations?hl=en&user=4_djMV0AAAAJ
> Research Gate
> https://www.researchgate.net/profile/Jinhui_Wang2/
>
|