# PROCESS macro or SEM for mediation model with latent variables

I am conducting a survey to analyze the effect of a binary variable X on a latent variable Y. The effect is assumed to be mediated by another latent variable M. Moreover, the effect of X on M is assumed to be moderated by a latent variable W.

My research on how to statistically examine this relationship led me to two different approaches:

1. Conditional Process Analysis with the SPSS macro PROCESS
2. A Structural Equation Model

Which one should I prefer and for what reasons?

I have read the following paper comparing the two approaches, but am still not able to make a decision: https://www.sciencedirect.com/science/article/abs/pii/S1441358217300265

Thank you very much for your help!

• If you have latents, use SEM. Jun 20 '19 at 17:16

Well, the PROCESS macro can't handle latent variables, and you have a latent variable, so I'm not sure why you'd even consider it when there is another clear option available. See Muthen & Asparouhov (2015) for an introduction to (causal) mediation with latent variables along with Mplus code. The sem function in Stata is also good for performing mediation analysis with latent variables.