James,
(1) The contrasts are the differentiated contrasts. There are n-1
contrasts where n is the number of levels. For the interaction, there
are (n-1)*(m-1) contrasts. These hold for all ANOVA models.
(2) The full factorial is not valid for repeated measure designs. The
statistics are inflated for all contrasts and inflated the most for
the group effects. This is due to the incorrect df and error terms. If
you use the flex. factorial model, then the error terms for trial
outcome and group*trial are correct, but the group effects are
inflated. Please see my previous posts on the issue.
(3) If you want to do interpret group, trial, group*trial effects in
the same model, then you need to use custom software. The group I am a
part of is in the process of developing the tools to do this and it
should be submitted for publication soon.
Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Postdoctoral Research Fellow, GRECC, Bedford VA
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
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 Thu, Sep 1, 2011 at 2:04 PM, Ellis, James <[log in to unmask]> wrote:
> Dear Experts,
>
>
>
> All these results pertain to a 4x3 ANOVA run on SST data, with two factors:
> Group – 4 levels, and Trial Outcome – 3 levels (1 = Corr Go, 2 = Corr Stop,
> 3 = Incorr Stop). However, the SPM-generated contrasts cut off one level
> per factor; see the list of contrasts below. Is this simply too many factor
> levels for SPM to consider?
>
>
>
> This is the list of contrasts offered:
>
>
>
> for i = 1:16;
>
> disp(SPM.xCon(i).name)
>
> end
>
> Average effect of condition
>
> Main effect of Group
>
> Main effect of Trial Outcome
>
> Interaction: Group x Trial Outcome
>
> Positive effect of condition_1
>
> Positive effect of Group_1
>
> Positive effect of Group_2
>
> Positive effect of Group_3
>
> Positive effect of Trial Outcome_1
>
> Positive effect of Trial Outcome_2
>
> Positive Interaction: Group x Trial Outcome_1
>
> Positive Interaction: Group x Trial Outcome_2
>
> Positive Interaction: Group x Trial Outcome_3
>
> Positive Interaction: Group x Trial Outcome_4
>
> Positive Interaction: Group x Trial Outcome_5
>
> Positive Interaction: Group x Trial Outcome_6
>
>
>
> Thanks,
>
> James
>
>
>
> Ps. The filelist is one of six pages just to give you an example. sF2 is
> group and sF3 is Trial Outcome
|