Hello! I am a bit new to STATA and I'm trying to do an analysis for my thesis. I used the SEM command to show there is a moderated mediation between those variables, as presented in the illustration. American Identity, perceived threat, and torture attitudes are ordinal variables. Discourse exposure is dichotomous with 0 = Low discourse exposure and 1 = High discourse exposure.
First, I simply ran the code for mediation:
sem (threat <- identity)(torture <- threat identity), vce(bootstrap,reps(5000))
estat teffects.
-> Obtained a coefficient for the indirect effect of identity on torture through the terrorist threat of about 0.05.
Then, I also used sem to show there is a moderated mediation in which exposure to political discourse affects path b between threat and torture attitudes:
sem (0: threat <- identity@b1 _cons@i1)(0: torture <- threat identity@b2) ///
(1: threat <- identity@b1 _cons@i1)(1: torture <- threat identity@b2), group(discourse) ///
variance(0: e.threat@v1 e.torture@v1) ///
variance(1: e.threat@v1 e.torture@v2)
estat teffects
-> And for the indirect effects in the moderated mediation, I got approximately 0.03 for low discourse exposure and 0.07 for high discourse exposure.
Now, what I wanted to do was to plot this moderated mediation. As mentioned, I am quite new to STATA and do not really know how to go about plotting (I am not sure what the graph is supposed to look like and hence, I do not know how to code it either - what components it should have and so on). I assume what I have to do is show the values of the indirect effect according to the two groups (low exposure and high exposure)...
If anyone more experience could help, I would be really really grateful! Thank you!
