Dear Doug,
I really appreciate all your answers, and to be sure that I understood the
reasoning for setting the contrast for the interaction effect.
Let's assume that we have 9 groups (two factor, with three levels each).
G1-left
G1-right
G1-ambi
G2-left
G2-right
G2-ambi
G3-left
G3-right
G3-ambi
[1 -1 0 -1 1 0 0 0 0]/2
[1 0 -1 -1 0 1 0 0 0]/2
[1 -1 0 0 0 0 -1 1 0]/2
[1 0 -1 0 0 0 -1 0 1]/2
However, if we have 8 groups (two factors, one of them with 4 levels)
G1-male
G1-female
G2-male
G2-female
G3-male
G3-female
G4-male
G4-female
[1 -1 -1 1 0 0 0 0]/2
[1 -1 0 0 -1 1 0 0]/2
[1 -1 0 0 0 0 -1 1]/2
Is this correct?
Thank you very much in advance,
Gabriel
Date: Wed, 9 Mar 2011 22:22:18 -0500
From: gr...@nmr.mgh.harvard.edu
To: gabriel...@hotmail.com; freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] glm question
two row contrast matrix
[1 1 -1 -1 0 0]
[1 1 0 0 -1 -1]/2
1st row test for a difference bet G1 and G2
2nd row tests for a difference bet G1 and G3
Note that you don't need to test G2 vs G3 (that is implied by the
other two tests).
doug
On 3/9/11 7:33 PM, Gabriel Go.Es. wrote:
Thanks a lot,
just one another question Doug,
how to set an ANOVA for group differences??'
Gabriel
> Date: Wed, 9 Mar 2011 17:52:03 -0500
> From: gr...@nmr.mgh.harvard.edu
> To: gabriel...@hotmail.com; freesurfer@nmr.mgh.harvard.edu
> Subject: Re: [Freesurfer] glm question
>
>
>
> Gabriel Go.Es. wrote:
> >
> > Thank you so much, Doug
> >
> > I've check at your update and saw that this contrast has
to be /2,
> > wich I guess means 0.5 instead of one to keep the scale
right?
> Yes.
> >
> > and the contrast for me, would be something like:
> > Let's suppose
> > group1_males
> > group1_females
> > group2_males
> > group2_females
> > group3_males
> > group3_females
> > for differences between males and females (males >
females)
> > regreessing the efects of group would be:
> > [0.333 -0.333 0.333 -0.333 0.333 -0.333]
> Correct
> > differences between group1 & group 3 regressing the
effects of gender:
> > [0.5 0.5 0 0 -0.5 -0.5]
> Correct
> > and the differences between males and females for a
single group can be:
> > [1 -1 0 0 0 0]
> > am I right???
> Yes
> > but I don't yet figure out how to set a difference from
0 in any of
> > the group means (to meassure the effect of group), nor
the interaction
> > between group and gender, and that's because my model is
not like the
> > one from the example,
> Try using
> [1 -1 -1 1 0 0]
> [1 0 0 0 -1 1]
>
> This is a single matrix with two rows. The first row tests
for an
> interaction between G1/2 and M/F. The second row tests for
G1/3 and M/F.
> Note that you don't need a third row.
>
> >
> >
> >
> >
> >
> > > Date: Wed, 9 Mar 2011 14:52:37 -0500
> > > From: gr...@nmr.mgh.harvard.edu
> > > To: gabriel...@hotmail.com
> > > CC: freesurfer@nmr.mgh.harvard.edu
> > > Subject: Re: [Freesurfer] glm question
> > >
> > > That contrast is wrong. I've updated this page with
the correct contrast
> > > https://surfer.nmr.mgh.harvard.edu/fswiki/Fsgdf6G0V
> > > which is:
> > >
> > > |+1 -1 0 -1 +1 0|
> > > |+1 0 -1 -1 0 +1|
> > >
> > > This is an F-contrast (two rows). The first row
looks for a difference
> > > between L and R for M and F (ie, a simple
interaction between LR and
> > > MF). The second row looks for a difference between
L and A for M and F.
> > > If either of these subcontrasts shows an effect,
then there is an
> > > interaction.
> > >
> > > doug
> > >
> > >
> > > Gabriel Go.Es. wrote:
> > > >
> > > > Dear FreeSurfer experts,
> > > >
> > > >
> > > >
> > > > I'm doing some group analyses and I'm a little
stocked with this:
> > > >
> > > > I have Six Groups (2 Factors, one with three
levels and the other one
> > > > with two levels), and one covariate. I don't
know the rule for
> > setting
> > > > the contrasts for an ANOVA and for the
interaction between the two
> > > > factors regressing out the covariate. I've
checked at the FS wiki and
> > > > found the following contrast for six groups
(two factors) and zero
> > > > covariates.
> > > >
> > > >
> > > >
> > > > Contrast 6 gender-x-handedness.mtx
(interaction between gender and
> > > > handedness)
> > > >
> > > > [0.333 -0.333 -0.333 0.333 0.000 0.000 0.000
0.000 0.000]
> > > >
> > > > [0.000 0.000 0.000 0.000 0.000 0.333 -0.333
-0.333 0.333]
> > > >
> > > >
> > > >
> > > > I don’t understand the rule that has been
followed.
> > > >
> > > > Can any one help me with this?
> > > >
> > > >
> > > >
> > > > Bests,
> > > >
> > > > Gabriel
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> >
------------------------------------------------------------------------
> > > >
> > > >
_______________________________________________
> > > > Freesurfer mailing list
> > > > Freesurfer@nmr.mgh.harvard.edu
> > > >
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
> > >
> > > --
> > > Douglas N. Greve, Ph.D.
> > > MGH-NMR Center
> > > gr...@nmr.mgh.harvard.edu
> > > Phone Number: 617-724-2358
> > > Fax: 617-726-7422
> > >
> > > Bugs:
surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
> > > FileDrop:
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
> > >
> > >
> > >
> > > 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.
> > >
>
> --
> Douglas N. Greve, Ph.D.
> MGH-NMR Center
> gr...@nmr.mgh.harvard.edu
> Phone Number: 617-724-2358
> Fax: 617-726-7422
>
> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
> FileDrop:
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
>
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
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.
