Hello,

I need to set up a factorial design for my resting state data and I am a bit confused about how to set this up in CONN. 

Essentially, I have 2 groups (patients-healthy), 2 conditions (sham-treatment), and 2 runs (pre-post). In this design, should I set my conditions to be just pre and post, specify sham and treatment in the 1st level, and patient and healthy groups in the 2nd level with dummy variables? Or do I need to use the conditions tab to encode sham and treatment together with pre and post (e.g. sham_pre, real_pre, sham_post, real_post), and use the 2nd level to specify groups? 

Thank you very much 

Hi,

It depends on whether your "conditions" (intervention/sham) is a within- or a between-subjects factor. Were those two conditions run for every participant, perhaps randomizing their order (that would make "conditions" a within-subject factor), or were subjects randomly assigned a single condition (i.e. each subject received either a sham or an intervention treatment; that would make "conditions" a between-subjects factor)?

In general all of the within-subject factors are defined in Setup.Conditions, and all of the between- subject factors are defined in Setup.Covariates (2nd-level). 

Hope this helps

Alfonso

Originally posted by ozzy:

Hello,

I need to set up a factorial design for my resting state data and I am a bit confused about how to set this up in CONN. 

Essentially, I have 2 groups (patients-healthy), 2 conditions (sham-treatment), and 2 runs (pre-post). In this design, should I set my conditions to be just pre and post, specify sham and treatment in the 1st level, and patient and healthy groups in the 2nd level with dummy variables? Or do I need to use the conditions tab to encode sham and treatment together with pre and post (e.g. sham_pre, real_pre, sham_post, real_post), and use the 2nd level to specify groups? 

Thank you very much