Two Dependent Variables:
Two Moderators:
It is possible to do this?
But in the diagram you attached here, Local of Awareness and Local of Satisfaction do' not seem to be moderators.
Hi you may take a look three path mediation model from a paper by Wilderom et all 2015 to deal with this. I think you can remove "government participation" or you you can utilize this as moderator variable.(https://www.researchgate.net/profile/Celeste_Wilderom/publication/274142185_From_manager's_emotional_intelligence_to_objective_store_performance_Through_store_cohesiveness_and_sales-directed_employee_behavior/links/552524b40cf2b123c5178be5.pdf)
Article From manager's emotional intelligence to objective store per...
Dear Ak,
you should multiply each IV by each moderator variable in SPSS database. Then, in Amos, you should design you model considering the following observed variables: IVs, Moderator variables, and each IVxModerator variable.
Imagine that X is the IV, M is the moderator nd Y is the DV. You should design a model in AMOS taking X, M, and XxM as IVs and Y as DV.
Just one small addition to Lisete's answer: mean-centre the X and M variables before hand. The easiest way to do that is to z-score these variables in SPSS before then multiplying them together and saving this as s new variable. Mean-centering reduces the problem of coliniarity between the X IV and X*IV variables in the subsequent models. See Tabachnick and Fidell "using multivariate statistics" for more.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling