Email discussion lists for the UK Education and Research communities

## SPM@JISCMAIL.AC.UK

#### View:

 Message: [ First | Previous | Next | Last ] By Topic: [ First | Previous | Next | Last ] By Author: [ First | Previous | Next | Last ] Font: Proportional Font

#### Options

Subject:

SwE with 3 time points analysis of 2 groups

From:

Date:

Thu, 2 Feb 2017 11:12:48 +0000

Content-Type:

text/plain

Parts/Attachments:

 text/plain (49 lines)
 ```Dear experts. I am using the SwE toolbox to analyze data of 2 groups (30 Subjects each) of volunteers acquired in 3 time points. I read the manual and the example analysis and tried to set the parameters: Scans: 180 files (Group1Time1(30), G1T2(30), G1T3(30), G2T1(30), G2T2(30), G2T3 (30)) Groups: kron(1:2,ones(1,90)) visits: kron(ones(1,60),1:3) Small sample adjustments: type 3 Degrees of freedom type: approx III Subject: kron(1:60,ones(1,3)) Design Matrix: "To check general differences between groups throughout the time points" Covariate 1      vector: ones(1,90)      name: G1 Covariate 2      vector: [zeros(1,90); ones(1,90)]      name: G2 Is this correct?? The contrast vector would be simplest in this case [1 -1]...right? "To check differences between groups time points" Covariate 1      vector: ones(1,30)      name: G1T1 Covariate 2      vector: [zeros(1,30); ones(1,30)]      name: G1T2 Covariate 3      vector: [zeros(1,60); ones(1,30)]      name: G1T3 Covariate 4      vector: [zeros(1,90); ones(1,30)]      name: G2T1 Covariate 5      vector: [zeros(1,120); ones(1,30)]      name: G2T2 Covariate 6      vector: [zeros(1,150); ones(1,90)]      name: G2T3 Is this correct?? I found a reply email form Prof. John Ashburner on SPM Archives and he mentioned a contrast image encoding rate of change between time points. Is this a GLM (like one sample ttest, including the 3 images and a covariate vector like [1 2 3] or [3 2 1], checking the correlation ([0 1]) to creates a con_? Thank you very much for any help! ```