Hello researchers,
I am going to implement social support in mediation model. However I can go for peers' related social support, teachers' related social support, or parental social support...
Iinitially, I could just apply those three dimensions in the mediation and see which one of them is significant (if any) for the direct effect.
Subsequently, I could use the most significant one for another mediation model where I would have other mediators which differ from social support, e.g. stress.
Is it a good idea to approach it like this? Or would it be better to include all three types of social support I mentioned along with stress and other mediators?
What are your opinions on that? What are the pros and cons of either way to approach that?
Thank you
Witold