If you want to test for interactions among nuisance variables, then you
have to create a new regressor where you multiply them together.
doug
On 3/30/14 4:20 PM, Clint Johns wrote:
I have a clarifying question about fsgd file format.
We have 40 participants, and a large battery of individual difference
measures (egg., vocabulary, phonological skill, etc.)
We applied a PCA to the battery and found 4 components, corresponding
to WMC, processing speed, comprehension ability, and phonological ability.
We tested for main effects of each of these on thickness, area, and
volume via QDEC.
However, we also want to check for interactions - and since this is
essentially 4 continuous factors, this is not QDEC-able (at least, it
does not appear to be). A nuisance variable (e.g., mean thickness) is
also present.
Nregressors = Nclasses*(Nvariables+1) = 1*((4 components + 1 nuisance)
+1) = 1*(5+1) = 6
Null contrast is
1 0 0 0 0 0
Main effect, e.g., of the working memory component (#1) is
0 1 0 0 0 0
Main effect, e.g., of the WM component modulated by the nuisance
variable is
0 1 0 0 0 1
So the format to examine possible interactions, e.g., component 1 and
component 2 interact to affect thickness, is
0 1 1 0 0 0
And with the nuisance variable
0 1 1 0 0 1
Do we have this right?
THANKS!!
Clint
--
I remain...
Clinton L. Johns, Ph.D.
Research Scientist, Haskins Laboratories
300 George Street
New Haven CT 06511
speech: 203-865-6163 x240
fax: 203-865-8963
net: [email protected] <mailto:[email protected]>
_______________________________________________
Freesurfer mailing list
[email protected]
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
[email protected]
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
