Dear Statalisters,
I am new to Stata and HLM and I would like to test a HLM model with moderated mediation.
All explanatory variables are at level 1 (M2LCO_C, M2ECS_C, M2EHS_C), and at level 2 we only want to take into account the Leader/supervisor (each supervisor rates M3EC for several employees/subordinates). As we determined through an ANOVA that there are supervisor/leader effects on M3EC, we have to control for the leader/supervisor at level 2. On the right part of the model (see figure) we have the control variables (EA & EG_D) and the moderating effects: EG_ECS (EG moderates the effect of M2ECS) & EG_EHS (EG moderates the effect of M2EHS), and so we have a moderated mediation model.
The code for this model is: . gsem (M2LCO_C -> M3EC, ) (M2LCO_C -> M2ECS_C, ) (M2LCO_C -> M2EHS_C, ) (M2ECS_C -> M3EC, ) (M2EHS_C -> M3EC, ) (EA -> M3EC, ) (EG_D -> M3EC, ) (M1[LEADER] -> M2ECS_C, ) (M3[LEADER] -> M3EC, ) (M2[LEADER] -> M2EHS_C, ) (EG_ECS -> M3EC, ) (EG_EHS -> M3EC, ), covstruct (_lexogenous, diagonal) latent(M1 M3 M2 ) nocapslatent
I am sure that the specification of the model needs to be improved and, thus, I would really appreciate your help!!
I am new to Stata and HLM and I would like to test a HLM model with moderated mediation.
All explanatory variables are at level 1 (M2LCO_C, M2ECS_C, M2EHS_C), and at level 2 we only want to take into account the Leader/supervisor (each supervisor rates M3EC for several employees/subordinates). As we determined through an ANOVA that there are supervisor/leader effects on M3EC, we have to control for the leader/supervisor at level 2. On the right part of the model (see figure) we have the control variables (EA & EG_D) and the moderating effects: EG_ECS (EG moderates the effect of M2ECS) & EG_EHS (EG moderates the effect of M2EHS), and so we have a moderated mediation model.
The code for this model is: . gsem (M2LCO_C -> M3EC, ) (M2LCO_C -> M2ECS_C, ) (M2LCO_C -> M2EHS_C, ) (M2ECS_C -> M3EC, ) (M2EHS_C -> M3EC, ) (EA -> M3EC, ) (EG_D -> M3EC, ) (M1[LEADER] -> M2ECS_C, ) (M3[LEADER] -> M3EC, ) (M2[LEADER] -> M2EHS_C, ) (EG_ECS -> M3EC, ) (EG_EHS -> M3EC, ), covstruct (_lexogenous, diagonal) latent(M1 M3 M2 ) nocapslatent
I am sure that the specification of the model needs to be improved and, thus, I would really appreciate your help!!
Attached Files
