I misread your reply in a hurry, which is why my response didn't make any sense. Sorry! After reading it again (carefully) I think I understand the contrast matrix. I do not understand why you have so many inputs for a 4 factor design with 2 subjects. I calculate 8 entries total (one input per subject for each factor. four factors*2 subjects = 8 inputs), but you have 16 entries. Are you assuming I have multiple runs for each factor? If this is the case, I would have assumes that I needed to model repeated measures of the same factor at the second level, then input their means on the third level. Am I mistaken?
One more question: If this design is essentially an ANOVA, to find out if there is any effect of my factor I only need to add an f-test and check off contrasts B-A, C-A, and D-A, yes?
Thanks Jeanette! As usual, you are extremely helpful.
Jen
Jennifer Bramen, Ph.D.
Assistant Research Neuroscientist
David Geffen School of Medicine at UCLA
Developmental Cognitive Neuroimaging
635 Charles Young Drive South
Los Angeles, CA 90095-7332
Phone: (310)267-5116
Fax: (310)525-0865
Email: [log in to unmask]
Campus Mail Code 176919
________________________________________
From: Bramen, Jennifer E.
Sent: 08 March 2009 13:52
To: FSL - FMRIB's Software Library
Subject: RE: [FSL] Quadruple Paired T-Test
I am doing a quadruple t-test, which is why there is an extra EV. The Triple T-Test is shown in the feat description on the FSL website, but I am not sure that I generalized it correctly. Also, this is a paired quadruple t-test. The last 2 EVs are basically the subject means. Does that help?
Thanks Jeanette.
Jen
Jennifer Bramen, Ph.D.
Assistant Research Neuroscientist
David Geffen School of Medicine at UCLA
Developmental Cognitive Neuroimaging
635 Charles Young Drive South
Los Angeles, CA 90095-7332
Phone: (310)267-5116
Fax: (310)525-0865
Email: [log in to unmask]
Campus Mail Code 176919
________________________________________
From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of Jeanette Mumford [[log in to unmask]]
Sent: 07 March 2009 14:42
To: [log in to unmask]
Subject: Re: [FSL] Quadruple Paired T-Test
Hi Jen,
I'm not exactly sure what your evs4-6 are, let me know if I'm misunderstanding something. The tripled test isn't as bad as it seems. It is just a 1 way anova with 3 levels using a factor effects approach. (ok, easy if you're familiar with factor effects).
Your first 3 evs look right
ev1=1 for level 1 and -1 for level 2, 0 for the rest
ev2=1 for level1 and -1 for level 3, 0 for the rest
ev3=1 for level1 and -1 for level4, 0 for the rest
then you should have as many remaining evs as subjects
ev4=1 for subject 1 and 0 otherwise
ev5=1 for subject 2 and 0 otherwise
ev6=1 for subject 3 and 0 otherwise
etc...
I have an easy trick for figuring out the contrasts with these ANOVA models (this is an ANOVA model, 1-way repeated measures). First here's the design for a couple of subjects. The first 8 entries are from sub1 and second 8 are sub2 and the level order goes a, a, b, b, c, c, d, d,
EV1 EV2 EV3 EV4 EV5
1 1 1 1 0
1 1 1 1 0
-1 0 0 1 0
-1 0 0 1 0
0 -1 0 1 0
0 -1 0 1 0
0 0 -1 1 0
0 0 -1 1 0
1 1 1 0 1
1 1 1 0 1
-1 0 0 0 1
-1 0 0 0 1
0 -1 0 0 1
0 -1 0 0 1
0 0 -1 0 1
0 0 -1 0 1
Ignore the subject-specific EVs and look across the rows corresponding to your 4 levels
So all 'a' rows look like the first row, 'b' rows are like the 3rd, etc.
a=ev1+ev2+ev3 -> [1 1 1 0 0 0] (use 0's for the subject-specific evs)
b=-ev1 -> [ -1 0 0 0 0 0]
c=-ev2 -> [0 -1 0 0 0 0]
d=-ev3 -> [0 0 0 -1 0 0]
a-b=[1 1 1 0 0 0]-[-1 0 0 0 0 0]=[2 1 1 0 0 0]
a-c=[1 1 1 0 0 0]-[0 -1 0 0 0 0]=[1 2 1 0 0 0]
a-d=[1 1 1 0 0 0]-[0 0 -1 0 0 0]=[1 1 2 0 0 0]
etc.
That ought to do the trick. Now you could even make up a quintuple paired t test if you wanted!
Cheers,
Jeanette
On Fri, Mar 6, 2009 at 10:53 AM, Bramen, Jennifer E. <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Dear List
I have to do a quadruple T-Test, and I know from the triple t-test example on your site that the design is counterintuitive. Can you check my design?
In this example, I have 2 subjects, scanned under 4 conditions.
A=a+b
B=-a
C=-b
D=-c?
EV1 EV2 EV3 EV4 EV5 EV6
1 1 1 1 1 0
1 1 1 1 0 1
-1 0 0 0 1 0
-1 0 0 0 0 1
0 -1 0 0 1 0
0 -1 0 0 0 1
0 0 -1 1 1 0
0 0 -1 1 0 1
B-A= (-a)-(a+b)
= -2a -b
C-A=-b - (-a) = -b+a
D-A = -c - (a+b) = -c -a -b
B-C = -a--b =(-a +b)
B-D=-b--c = -a +c
EV1 EV2 EV3 EV4 EV5 EV6
B-A -2 -1 0 0 0 0
C-A 1 -1 0 0 0 0
D-A -1 -1 -1 0 0 0
B-C -1 1 0 0 0 0
B-D -1 0 1 0 0 0
Is this correct?
Thanks
Jen
Jennifer Bramen, Ph.D.
Assistant Research Neuroscientist
David Geffen School of Medicine at UCLA
Developmental Cognitive Neuroimaging
635 Charles Young Drive South
Los Angeles, CA 90095-7332
Phone: (310)267-5116
Fax: (310)525-0865
Email: [log in to unmask]<mailto:[log in to unmask]>
Campus Mail Code 176919
________________________________________
IMPORTANT WARNING: This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential. You, the recipient, are obligated to maintain it in a safe, secure and confidential manner. Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the intended recipient, please immediately notify us by return email, and delete this message from your computer.
IMPORTANT WARNING: This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential. You, the recipient, are obligated to maintain it in a safe, secure and confidential manner. Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the intended recipient, please immediately notify us by return email, and delete this message from your computer.
|