Which test?

#1
Hello everyone,
A question that probarly seems to be a very simple question for many of you, but I’m a bit of a statistics newby: I have a research design with a 0-measurement (the questions: how often do you exercise per week and how often have you been exercising in the past month) an intervention (leaflets and a video about the importance of regular exercise and the health disadvantages if someone does not exercise enough) and then a follow-up one month later (same questions as the 0 measurement). Of course with the aim of measuring whether participants have started exercising more often after the intervention (leaflets and video). Do I then only have to do a paired t-test to measure the effect of the intervention? Or should I opt for an ANCOVA and, for example, see if age and gender play a role?
Thank you all in advance for thinking along!
 

Karabiner

TS Contributor
#2
(the questions: how often do you exercise per week and how often have you been exercising in the past month)
What was the answering format? Did participants just fill in a number,
or were there pre-deifined categories (such as "0 / 1-3 / 4-10 / > 10x")?
Of course with the aim of measuring whether participants have started exercising more often after the intervention (leaflets and video).
Please report group sizes and how subects were allocated to groups
(by randomization?) and how many subjects were lost to follow-up in
each group.
Do I then only have to do a paired t-test to measure the effect of the intervention? Or should I opt for an ANCOVA and, for example, see if age and gender play a role?
What exactely do you want? Analyse just the effect of intervention group vs. control
group? Or do you rellay want to include additional variables? If yes, then what exactly
do you mean by "role" - moderators? mediators? confounders (e.g. in case of non-
random allocation)?

With kind regards

Karabiner