Print

Print
Choosing any 2 of these will create the F-contrast and will produce the same results. Generally, I like choosing (1) and (3).

Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Postdoctoral Research Fellow, GRECC, Bedford VA
Website: http://www.martinos.org/~mclaren
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
intended only for the use of the individual or entity named above. If the
reader of the e-mail is not the intended recipient or the employee or agent
responsible for delivering it to the intended recipient, you are hereby
notified that you are in possession of confidential and privileged
information. Any unauthorized use, disclosure, copying or the taking of any
action in reliance on the contents of this information is strictly
prohibited and may be unlawful. If you have received this e-mail
unintentionally, please immediately notify the sender via telephone at (773)
406-2464 or email.


On Mon, Apr 29, 2013 at 7:38 PM, Maren Strenziok <[log in to unmask]> wrote:

Hi all,


I have a question regarding setting up F tests for a 3x2 ANOVA. My factors are time (pre- and post- intervention) and group (intervention 1, 2 and 3). I am only interested in the group x time interaction effect. Someone explained previously on this list that a good way of doing this is to enter post-pre intervention (or pre-post-intervention) difference maps as input data. This worked well, technically, but I have no a prior hypothesis about the direction of the effect (i.e., I don't know whether I will see an increase or a decrease in my measure – BOLD activation at rest). So I set the model up again entering the pre- and post-intervention maps separately (rather than their differences). Below is the model that I set up (example with 2 subjects in each group; the first inputs are pre-intervention maps, the last inputs are post-intervention maps). I then went on to stting up the F tests and got an error message. According to the balloon help, I can set up three contrasts (A >B, A > C, and B > C, where A, B, and C are my three intervention groups). This worked fine, but when I select an F test to look at the overall effect of treatment across all three groups (see below), I get the following error message:

"problem with processing the model: F test 1 isn't valid – each included contrast cannot be a linear combination of the others"

Can I look at the overall effect of the three groups at all, and if so, how do I set this up?


Group   A   B   C   EV4   EV5    EV6   EV7   EV8   EV9

1           1   0    0       1       0         0       0        0       0

1           1   0    0       0       1         0       0        0       0

2           0   1    0       0       0         1       0        0       0

2           0   1    0       0       0         0       1        0       0

3           0   0    1       0       0         0       0        1       0

3           0   0    1       0       0         0       0        0       1

1          -1   0    0       1       0         0       0        0       0

1          -1   0    0       0       1         0       0        0       0

2           0  -1    0       0       0         1       0        0       0

2           0  -1    0       0       0         0       1        0       0

3           0   0   -1       0       0         0       0        1       0

3           0   0   -1       0       0         0       0        0       1


                                                                                     F test

A>B      1   -1    0       0       0         0       0        0       0      x

A>C      1    0   -1       0       0         0       0        0       0      x

B>C      0    1   -1       0       0         0       0        0       0      x