Which statistical test? I've shown my attempt to work it out.

#1
2 Questionnaires (for hope and self-image) are given to participants before, during, and after an intervention.

All participants take part in the intervention to see if it has any immediate/delayed impact on depression and self-image levels of participants.

I know it's within subjects. I think there are 2 continuous DVs (hope/self image).

I'm confused if there are 1 or 2 IVs? Is the IV just time or is it also whether or not they've had the intervention yet too?

So I'm not sure if this should be a two-way repeated measures ANOVA or a MANOVA, or if I've just gotten myself confused somewhere.

Thank you!
 

CowboyBear

Super Moderator
#2
All participants take part in the intervention to see if it has any immediate/delayed impact on depression and self-image levels of participants.
Sorry to be the bearer of bad news, but if I'm understanding your design correctly this study is not capable of telling you whether or not the intervention has a causal impact on the DV's. Scores on the DV may change after the intervention, but you won't be able to conclude that this was due to the intervention. A control group with random assignment to treatment/control would allow you to make causal inferences.

Is the IV just time
Yep

So I'm not sure if this should be a two-way repeated measures ANOVA or a MANOVA, or if I've just gotten myself confused somewhere.
You could use two separate repeated measures ANOVA (depression on time and self-image on time). A MANOVA approach could work too I guess but studying impacts on the two DVs separately is probably closer to what you're actually trying to find out.
 
#3
Thank you, that is very helpful!

Would I need to do anything else for the control condition (other than give the questionnaire before/after)? I think I'm supposed to just leave the control group and not do anything with them but I'm not sure if that is right.

Thank you!
 

Karabiner

TS Contributor
#4
It depends on how the intervention (the active ingredients) is conceptualized, and what exactly you want to determine. For example, offering an intervention could have some impact on depression anyway, regardless of what exactely you offer, just because people become involved/activated/hopeful etc. (in pharmaceutical trials, one offers a placebo instead of just doing nothing). So maybe some kind of "active" control condition could be necessary, if it is intended to study whether the intervention has an impactbeyond regression effects and placebo effects. Or at least one could use a waiting list control group. There's a lot of studies on the evaluation of depression treatments/interventions, so you could do some research how control groups are defined there.

What surprises me a bit, you perform experiments with patients without having knowledge about the design of such experiments, how (or where) is that possible?

With kind regards

K.