Hello. I'm working on a model with 1 predictor, 2 mediations and 2 outcomes in SPSS AMOS. I want to see the relationship between 2 mediators but I can't draw the covariance arrow for them (It seems that it just works for predictors). Is there any way to do that?
In AMOS, first you have to ensure measurement model which confirms ur validity and reliability as well as model fit. If you are unaware about measurement model u can watch videos. When you confirm MM, the next model is structure model where you specify your mediator and IVs, DVs etc. But in SM you also see for model fitness.
Thank you for your answer. So, what should I do is that, besides the results of my model if I want to indicated the relationship of mediator variables, I need to write the output o the measurement model, right?
Why do you want to see the relationship between the two variables through SEM? Can't you just run a correlation to see how the two mediator variables correlate?
My mentor asked me to do this in order to see if the relation of mediators has an effect on the model. Apparently in some other statistic programmes for SEM the covariation arrow can be drawn.
If you have a theoretical causal proposition between the two (eg. mediator A leads to mediator B) then you can draw a directional line from A to B. If not, you can covary the error terms/residuals of the two mediators. What this means is that there is some common cause(s) unaccounted for in your model. Important to do this only with theoretical reasons, as it can erroneously inflate your model fit indices.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS