help
help > RE: Displaying Results in Calculator
Jul 22, 2021 03:07 PM | Alfonso Nieto-Castanon - Boston University
RE: Displaying Results in Calculator
Hi Nick
Yes, that is exactly correct, and the scatterplot represents the association between Frailty (on the x-axis) and functional connectivity (connectivity between Hippocampus and that individual cluster; on the y-axis).
Best
Alfonso
ps. Just for general reference, on that plot red-dots are the actual data, and blue-dots are the fitted data (how well could you predict connectivity knowing a subject's age, sex, education, and frailty scores), so the effect of frailty is in that plot roughly characterized by the approximate "slope" of the cloud of blue dots. Other interesting info that can be extracted from these plots is that the difference between a pure straight-line fit to that cloud and the actual blue-dots position represents the size of the effect of the other covariates (age/sex/education) in the model. Last, if you right-click on that plot and select the 'show null-hypothesis' option, that will also show you some additional gray-dots representing how well could you predict connectivity knowing a subjects' age, sex and education (without knowing its frailty scores), so the difference between the gray and blue dots quantifies the size of the "frailty effect" that your statistical analysis are actually evaluating
Originally posted by Nick Bray:
Yes, that is exactly correct, and the scatterplot represents the association between Frailty (on the x-axis) and functional connectivity (connectivity between Hippocampus and that individual cluster; on the y-axis).
Best
Alfonso
ps. Just for general reference, on that plot red-dots are the actual data, and blue-dots are the fitted data (how well could you predict connectivity knowing a subject's age, sex, education, and frailty scores), so the effect of frailty is in that plot roughly characterized by the approximate "slope" of the cloud of blue dots. Other interesting info that can be extracted from these plots is that the difference between a pure straight-line fit to that cloud and the actual blue-dots position represents the size of the effect of the other covariates (age/sex/education) in the model. Last, if you right-click on that plot and select the 'show null-hypothesis' option, that will also show you some additional gray-dots representing how well could you predict connectivity knowing a subjects' age, sex and education (without knowing its frailty scores), so the difference between the gray and blue dots quantifies the size of the "frailty effect" that your statistical analysis are actually evaluating
Originally posted by Nick Bray:
Hi Alfonso,
My project is exploring the correlation between Frailty and Connectivity.
My between-subject contrast is:
Effect of Frailty, controlling for age, sex, and education. Set-up as:
[0 1 0 0 0]
[AllSubjects Frailty Age Sex Education]
I conducted a seed-voxel analysis using the right Hippocampus, and it demonstrated higher or increasing connectivity with one cluster. Meaning, increasing connectivity between the Hippocampus and cluster is correlated with increasing Frailty. I now want to display this in the calculator for a visual representation:
1) I "Imported Values" of the cluster showing increased connectivity with the Hippocampus to my 2nd-level covariates. Then in calculator I...
2) Set up my between-subjects contrast in an identical format:
Effect of Frailty, controlling for age, sex, and education. Set-up as:
[0 1 0 0 0]
[AllSubjects Frailty Age Sex Education]
3) For between-measures contrast, I simply selected the imported cluster. *See Screenshot further clarification*
In short, is this approach correct? Is the scatterplot a visual representation of my 2nd-level results? I reviewed this thread (https://www.nitrc.org/forum/message.php?...) and believe I am on the track but just wish to confirm.
Cheers,
Nick
My project is exploring the correlation between Frailty and Connectivity.
My between-subject contrast is:
Effect of Frailty, controlling for age, sex, and education. Set-up as:
[0 1 0 0 0]
[AllSubjects Frailty Age Sex Education]
I conducted a seed-voxel analysis using the right Hippocampus, and it demonstrated higher or increasing connectivity with one cluster. Meaning, increasing connectivity between the Hippocampus and cluster is correlated with increasing Frailty. I now want to display this in the calculator for a visual representation:
1) I "Imported Values" of the cluster showing increased connectivity with the Hippocampus to my 2nd-level covariates. Then in calculator I...
2) Set up my between-subjects contrast in an identical format:
Effect of Frailty, controlling for age, sex, and education. Set-up as:
[0 1 0 0 0]
[AllSubjects Frailty Age Sex Education]
3) For between-measures contrast, I simply selected the imported cluster. *See Screenshot further clarification*
In short, is this approach correct? Is the scatterplot a visual representation of my 2nd-level results? I reviewed this thread (https://www.nitrc.org/forum/message.php?...) and believe I am on the track but just wish to confirm.
Cheers,
Nick
Threaded View
| Title | Author | Date |
|---|---|---|
| Nick Bray | Jul 22, 2021 | |
| Alfonso Nieto-Castanon | Jul 22, 2021 | |
| Nick Bray | Jul 22, 2021 | |