help
help > RE: ANCOVA with 3 groups
Apr 6, 2015 03:04 AM | Alfonso Nieto-Castanon - Boston University
RE: ANCOVA with 3 groups
Dear Liron,
You should select all of your second-level regressors (e.g. 'Group1', 'Group2', 'Group3', 'Age', 'Gender','other1', and 'other2') and enter a contrast [1 -1 0 0 0 0 0; 0 1 -1 0 0 0 0] in order to look at any between-group differences (between the first three selected subject-effects) after controlling for your 4 covariates of no interest (the next four selected subject-effects)
Let me know if you would like me to further clarify
Best
Alfonso
Originally posted by Liron Bensky:
You should select all of your second-level regressors (e.g. 'Group1', 'Group2', 'Group3', 'Age', 'Gender','other1', and 'other2') and enter a contrast [1 -1 0 0 0 0 0; 0 1 -1 0 0 0 0] in order to look at any between-group differences (between the first three selected subject-effects) after controlling for your 4 covariates of no interest (the next four selected subject-effects)
Let me know if you would like me to further clarify
Best
Alfonso
Originally posted by Liron Bensky:
Dear Alfonso
I want to run an ANCOVA, and can't seem to find the correct contrast: I have 3 groups (two patient groups and one control group), and 4 second-level covariates (age, gender ect'). I want to know if there's any differences between the groups, after controlling for all covariatres at once (i.e. a single model, that I'll later enter into a post-hoc comparisons for group effects, if anything comes out significant). I entered both the groups and the covariates as 'second level covariates' (is that the correct way?), and I'm trying to figure out the contrast that I need to use. I couldn't figure it out from the manual or the forum... Any suggestion that you may have would be greatly appreciated!
Thank you very much!!
Liron.
I want to run an ANCOVA, and can't seem to find the correct contrast: I have 3 groups (two patient groups and one control group), and 4 second-level covariates (age, gender ect'). I want to know if there's any differences between the groups, after controlling for all covariatres at once (i.e. a single model, that I'll later enter into a post-hoc comparisons for group effects, if anything comes out significant). I entered both the groups and the covariates as 'second level covariates' (is that the correct way?), and I'm trying to figure out the contrast that I need to use. I couldn't figure it out from the manual or the forum... Any suggestion that you may have would be greatly appreciated!
Thank you very much!!
Liron.
Threaded View
| Title | Author | Date |
|---|---|---|
| Liron Bensky | Mar 30, 2015 | |
| Alfonso Nieto-Castanon | Apr 6, 2015 | |
| Liron Bensky | Apr 6, 2015 | |
| Alfonso Nieto-Castanon | Apr 6, 2015 | |
| Liron Bensky | Apr 6, 2015 | |
| Fred Uquillas | Apr 7, 2015 | |
| Alfonso Nieto-Castanon | Apr 7, 2015 | |
| Liron Bensky | Apr 13, 2015 | |