Which statistical test I have to use for pre/post test?

#1
Dear readers,

I am wirting my thesis and I would like to test the effects of an mindfulness Intervention (independent variable) on reducing depressive, anxiety and stress symptoms (3 dependent variables). I have a pre and post-test for this and I would also like to include a control group.
Which test do I have to use for this and how can I calculate my power to knwo how much particpants I have to include for this study?

Thank you in advance
Kind regards,
Naomi
 

Karabiner

TS Contributor
#2
One common approach would be "mixed" analysis of variance, i.e.
repeated-measures analysis of variance with a repeated-measures
factor "time of measurement" (pre vs. post) and an additional
between-subjects (group) factor.

You could do this for each dependent variable separately. I am
not sure whether it is possible to include all 3 dependent measures
at the same time, as a second repeated-measures factor ("type of
response"). Alternatively you could perhaps the 3 measurements
into one "general distress" measure; the overlap between those 3
domains is huge, AFAIK.

With kind regards

Karabiner
 
#3
One common approach would be "mixed" analysis of variance, i.e.
repeated-measures analysis of variance with a repeated-measures
factor "time of measurement" (pre vs. post) and an additional
between-subjects (group) factor.

You could do this for each dependent variable separately. I am
not sure whether it is possible to include all 3 dependent measures
at the same time, as a second repeated-measures factor ("type of
response"). Alternatively you could perhaps the 3 measurements
into one "general distress" measure; the overlap between those 3
domains is huge, AFAIK.

With kind regards

Karabiner
Thank you very much for you quick reply.
The thing is, that its research question only has one independent variable (mindfulness) and one dependendent (distress), with a control group. The dependent variable consist of subfacets. If this is the case, can I do the same design without the repeated measures?
Kind regards
 

Karabiner

TS Contributor
#4
If you have 1 dependent variable distress, which is composed of anxiety, depression and stress,
then a mixed ANOVA with the repeated measures factor pre/post and the group factor looks
plausible.

With kind regards

Karabiner
 

j58

Active Member
#5
How about computing the pre–post difference for each subject, and then comparing the mean pre–post difference for the intervention and control groups using a 2-sample t-test.