Email discussion lists for the UK Education and Research communities

## FSL@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:

Re: GLM for 3 (between subjects) x 2 (within subjects) repeated measures ANOVA

From:

Date:

Thu, 11 Oct 2018 18:09:30 +0100

Content-Type:

multipart/mixed

Parts/Attachments:

 text/plain (35 lines) , GLM for 3 x 2 repeated measures ANOVA.png (35 lines)
 ```I appreciate looking at previous posts that this GLM may not be possible due to problems with assumptions of sphericity in mixed models. However I saw this previous posts from Anderson Winkler: https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=FSL;a3eedcc7.1501 See his attached GLM model. In this model have I concluded the following correctly? - A B C could be 3 groups of subjects and coding within EV1-3 is 1 or -1 could be for drug or placebo, and EVs 4-N are pairing subject EVs. - C1-3 would then be effect of drug separately in groups A to C, while C4-6 is differential effect of drug in groups A vs B, A vs C and B vs C . - F1 test for C4-5 would be overall drug x group interaction. Can one therefore add C7 as 0.33 0.33 0.33 0 0 0 …… as a contrast for overall effect of drug independent of group (i.e. collapsed across A B and C)? I note from comments by Anderson that in this GLM set up the 1 1 2 2 3 3 in first column is for randomise as it represents exchangeability blocks, but for standard Feat would replace these with all 1 1 1 1 1 1…..(just one big group, as that represents then variance groups). In randomise he also suggested the following instructions: - For contrasts C1-C3, use the option "-e design.grp" in randomise. - For contrasts C4-C6, and for the F-test, use the option "-e design.grp" and also the option "--permuteBlocks" in randomise. You may need to split the contrast file (*.con) into two, one with just C1-C3, another with just C4-C6, then make the necessary changes to the *.fts file (F-contrast), and run randomise twice (in separate directories to avoid overwriting results). Thanks Tony ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 ```