help
help > RE: Questions about ANOVA models
Oct 1, 2014 06:10 PM | Alfonso Nieto-Castanon - Boston University
RE: Questions about ANOVA models
Hi Jeff
Some thoughts on your questions below.
Best
Alfonso
Originally posted by Jeff Browndyke:
If you have scans for 10 subjects (the first five are patients and the last five controls) then the covariate "patients" would be defined as [1 1 1 1 1 0 0 0 0 0] and the covariate "controls" would be defined as [0 0 0 0 0 1 1 1 1 1].
If your five patients were scanned twice (and you entered the data in the order: first five "pre" scans for your patients, then five "post" scans for your patients, and last five scans for your controls) then the "patients-pre" covariate would be [1 1 1 1 1 0 0 0 0 0 0 0 0 0 0], the "patients-post" covariate would be [0 0 0 0 0 1 1 1 1 1 0 0 0 0 0] and the "controls" covariate would be [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1].
Hope this helps
Alfonso
Some thoughts on your questions below.
Best
Alfonso
Originally posted by Jeff Browndyke:
Alfonso,
Thank you for pointing this prior post out to me. I'm new to CONN.
Just two questions, however, how does one do the following in your instruction below, "define the two subject groups (e.g., patient and controls) as two simple dummy-coded second-level covariates in Setup.Covariates.Second-level"?
Thank you for pointing this prior post out to me. I'm new to CONN.
Just two questions, however, how does one do the following in your instruction below, "define the two subject groups (e.g., patient and controls) as two simple dummy-coded second-level covariates in Setup.Covariates.Second-level"?
If you have scans for 10 subjects (the first five are patients and the last five controls) then the covariate "patients" would be defined as [1 1 1 1 1 0 0 0 0 0] and the covariate "controls" would be defined as [0 0 0 0 0 1 1 1 1 1].
Also, in the case of repeated measures for
patients but a single timepoint scan for controls, how does one do
the following in your instruction, "define three second-level
covariates in Setup.Conditions defining these three subject groups
(e.g., patients-pre, patients-post and controls, dummy coded using
1/0 values identifying the corresponding scans for each group)."
How would that dummy coding look for three groups?
If your five patients were scanned twice (and you entered the data in the order: first five "pre" scans for your patients, then five "post" scans for your patients, and last five scans for your controls) then the "patients-pre" covariate would be [1 1 1 1 1 0 0 0 0 0 0 0 0 0 0], the "patients-post" covariate would be [0 0 0 0 0 1 1 1 1 1 0 0 0 0 0] and the "controls" covariate would be [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1].
Hope this helps
Alfonso
Threaded View
| Title | Author | Date |
|---|---|---|
| Yong Li | Sep 15, 2014 | |
| Alfonso Nieto-Castanon | Sep 17, 2014 | |
| Yong Li | Feb 13, 2015 | |
| Alfonso Nieto-Castanon | Feb 18, 2015 | |
| Yong Li | Feb 24, 2015 | |
| Jeff Browndyke | Oct 1, 2014 | |
| Alfonso Nieto-Castanon | Oct 1, 2014 | |
| Traute Demirakca | Oct 2, 2014 | |
| Jeff Browndyke | Oct 1, 2014 | |