[color=#000000]Dear Lydia,[/color]
[color=#000000]Regarding (1) yes, that is exactly correct[/color]
[color=#000000]Regarding (2) you may use the 'contrast manager' menu to save those individual contrasts once, and then you will be able to just select those saved contrasts each time[/color]
[color=#000000]Regarding (3), yes, definitely, analyses of main effects aggregating across multiple levels of one or several other factors are only meaningful when there are no interactions, in all other cases you typically want to break down those main effects into "simple main effects" at each level of the (potentially interacting) other factors (in your example, yes, you may simply select the first four conditions and enter a [1 -1 -1 1] contrast to indicate a B*C interaction test at the first-level of your A factor)[/color]
[color=#000000]Hope this helps[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by Lydia Vinals:[/i][quote]Dear all,
I am running a task-dependent functional connectivity analysis using gPPI. I have a 2 x 2 x 2 factorial design in an event-related paradigm. I have entered my condition-specific onset times (8 conditions) and ran a first-level analysis. I just have a question regarding my second level analysis, and more specifically, about testing for interactions between my conditions. On the second-level analysis tab, I have a list of my 8 conditions. In between-condition contrasts, I have tested for main effects by entering vectors as follows:
ME of condition A: [1 1 1 1 -1 -1 -1 -1]
ME of condition B: [1 1 -1 -1 1 1 -1 -1]
ME of condition C: [1 -1 1 -1 1 -1 1 -1]
I then want to test for 2-level interactions (e.g. INT condition A and B = [1 1 -1 -1 -1 -1 1 1]) and for the 3-way interactions between all conditions ([1 -1 -1 1 -1 1 1 -1].
My questions are:
1. Is this the correct way to be testing for main effects and interactions by manually entering between-condition contrasts?
2. Is there a way to import a matrix of contrasts to save me from having to enter them manually each time?
3. How can I interpret/break down interactions, particularly higher level ones? Can I simply look at a subset of the conditions? For example, looking for the interaction between condition B and C for one level of condition [1 -1 -1 1]?
Many thanks for your help![/quote]
[color=#000000]Regarding (1) yes, that is exactly correct[/color]
[color=#000000]Regarding (2) you may use the 'contrast manager' menu to save those individual contrasts once, and then you will be able to just select those saved contrasts each time[/color]
[color=#000000]Regarding (3), yes, definitely, analyses of main effects aggregating across multiple levels of one or several other factors are only meaningful when there are no interactions, in all other cases you typically want to break down those main effects into "simple main effects" at each level of the (potentially interacting) other factors (in your example, yes, you may simply select the first four conditions and enter a [1 -1 -1 1] contrast to indicate a B*C interaction test at the first-level of your A factor)[/color]
[color=#000000]Hope this helps[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by Lydia Vinals:[/i][quote]Dear all,
I am running a task-dependent functional connectivity analysis using gPPI. I have a 2 x 2 x 2 factorial design in an event-related paradigm. I have entered my condition-specific onset times (8 conditions) and ran a first-level analysis. I just have a question regarding my second level analysis, and more specifically, about testing for interactions between my conditions. On the second-level analysis tab, I have a list of my 8 conditions. In between-condition contrasts, I have tested for main effects by entering vectors as follows:
ME of condition A: [1 1 1 1 -1 -1 -1 -1]
ME of condition B: [1 1 -1 -1 1 1 -1 -1]
ME of condition C: [1 -1 1 -1 1 -1 1 -1]
I then want to test for 2-level interactions (e.g. INT condition A and B = [1 1 -1 -1 -1 -1 1 1]) and for the 3-way interactions between all conditions ([1 -1 -1 1 -1 1 1 -1].
My questions are:
1. Is this the correct way to be testing for main effects and interactions by manually entering between-condition contrasts?
2. Is there a way to import a matrix of contrasts to save me from having to enter them manually each time?
3. How can I interpret/break down interactions, particularly higher level ones? Can I simply look at a subset of the conditions? For example, looking for the interaction between condition B and C for one level of condition [1 -1 -1 1]?
Many thanks for your help![/quote]