help
help > RE: T value interpretation in 2nd level analysis
Aug 30, 2016 05:08 PM | Alfonso Nieto-Castanon - Boston University
RE: T value interpretation in 2nd level analysis
Hi Magno,
I am assuming here that your second-level anaysis is defining a regression model (e.g. selecting 'AllSubjects' and 'myCovariate' in the between-subject effects list, and entering a [0 1] between-subjects contrast in the between-subject contrast field). If that is the case, then the T statistic reported can be interpreted as a ratio between the variability in connectivity values covarying with 'myCovariate' divided by the remaining/residual variability (you can think of the two-sample t-test case comparing two groups of subjects simply as a specific subcase of this general regression model where your covariate is a step function regressor -e.g. 0's for one group and 1's for the other group's subjects-).
Hope this helps
Alfonso
Originally posted by Magno Guillen:
I am assuming here that your second-level anaysis is defining a regression model (e.g. selecting 'AllSubjects' and 'myCovariate' in the between-subject effects list, and entering a [0 1] between-subjects contrast in the between-subject contrast field). If that is the case, then the T statistic reported can be interpreted as a ratio between the variability in connectivity values covarying with 'myCovariate' divided by the remaining/residual variability (you can think of the two-sample t-test case comparing two groups of subjects simply as a specific subcase of this general regression model where your covariate is a step function regressor -e.g. 0's for one group and 1's for the other group's subjects-).
Hope this helps
Alfonso
Originally posted by Magno Guillen:
Hi Conn users
I just rephrase my question. Can anybody explain to me the meaning of the T value in the second level analysis results? It is generally assumed that T value is a measure of the difference between the means of 2 groups. which would be these two groups in our case?
Any insight is deeply appreciated
Magno
I just rephrase my question. Can anybody explain to me the meaning of the T value in the second level analysis results? It is generally assumed that T value is a measure of the difference between the means of 2 groups. which would be these two groups in our case?
Any insight is deeply appreciated
Magno
Threaded View
| Title | Author | Date |
|---|---|---|
| Magno Guillen | Aug 24, 2016 | |
| Magno Guillen | Aug 29, 2016 | |
| Alfonso Nieto-Castanon | Aug 30, 2016 | |
| Magno Guillen | Sep 1, 2016 | |
| Magno Guillen | Sep 1, 2016 | |