Hi all!
For my master's thesis, I am investigating the effects of movement on an exposure intervention for PTSD symptoms. PTSD symptoms were measured at pretest and posttest, and the participants were divided into 4 groups: Control, stationary exposure, exposure with acute movement and exposure with delayed movement.
For the analysis, I want to include the participants' movement habits as a moderator, and measure the differences in symptom reduction for the conditions that include movement, and those that don't.
I've been looking for an efficient way of conducting this analysis, and came across the MEMORE macro for SPSS, but I've also been thinking of attempting this with repeated measures ANOVA.
Which approach do you think would be more useful and efficient for the analysis?
Thanks in advance!
I had never heard of the MEMORE macro, but I see that it is based on an article by Montoya & Hayes (2017).
- https://www.akmontoya.com/spss-and-sas-macros
It sounds like you have an experimental design with (pseudo) random allocation to groups. If so, I would use ANCOVA with the baseline score as the covariate and the follow-up score as the DV. See this article, for example:
Article Statistics Notes: Analysing controlled trials with baseline ...
In SPSS, you can estimate that model via the UNIANOVA command (or the equivalent GLM command).
You said you want to "measure the differences in symptom reduction for the conditions that include movement, and those that don't." Does that mean you want to compare [Control, Exposure-Stationary] to [Exposure-Acute, Exposure-Delayed]? If so, one way to get that contrast, I think, would be to include a Helmert contrast. Something like this:
/CONTRAST(Group)=HELMERT
Or possibly:
/CONTRAST(Group*MoveHabits)=HELMERT
Another option is to use LMATRIX to get exactly the contrasts you want, but that is trickier, and it is easy to get things wrong.
HTH.
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Powders