I have two groups of people (controls and subjects) that are scanned before and after a class. Each person is scanned twice, two months apart. I am interested in seeing how the intervention changed the activation of the subjects relative to the controls. Mathematically, I want (Sub_after – Sub_before) – (Control_after – Control_before). How would I set that up in second level analysis?
Currently I have the second level analysis set up like this (the first level analysis is the difference between them doing a computation task vs. a counting task):
Group EV1(Control_before) EV2 (Sub_before) EV3(Control_after) EV4 (Sub_after) Scanning Session Number
1 1 0 0 0 1
1 1 0 0 0 2
1 1 0 0 0 3
1 1 0 0 0 4
1 1 0 0 0 5
1 0 1 0 0 6
1 0 1 0 0 7
1 0 0 1 0 8
1 0 0 1 0 9
1 0 0 1 0 10
1 0 0 1 0 11
1 0 0 1 0 12
1 0 0 0 1 13
1 0 0 0 1 14
I have 5 controls and 2 subjects (I am actually scanning more subjects, but if I can do it here, I can do it for all the subjects). Scanning sessions 1-5 are the controls before the intervention. Scanning sessions 6&7 are the subjects before the intervention. Scanning sessions 8-12 are the controls after the intervention (after the same amount of time, because they did not get the intervention). Scanning sessions 13&14 are the subjects after the intervention.
I want to know how much activation in the subjects changed due to the intervention relative to how much it changed in the controls over the same amount of time. The contrast is:
EV1(Control_before) EV2 (Sub_before) EV3(Control_after) EV4 (Sub_after)
1 -1 -1 1
Mathematically, this is (Sub_after – Sub_before) – (Control_after – Control_before) = Sub_after + Control_before - Sub_before - Control_after
Does this make sense?
Eric Walden
James C. Wetherbe Associate Professor
Rawls College of Business
Texas Tech University