Specification in SEM builder for Multilevel SEM

[Kevin Nanakdewa]

Hello, I'm using Stata 15. Trying to figure out how to specify my model using SEM Builder. I can run a simple multilevel mediation model using the GSEM mode but can't find a good example online of a more complex model. In particular, I'm not sure where I need to add multilevel latent variables. The network characteristic variables are computed based on the network that the ego reports. The behaviors and interpersonal consequences are all 5-item measures reported by the ego about each alter. Thanks in advance for your help, let me know if any additional information is needed.

[Phil Bromiley]

Welcome to Statalist. You will increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

I don't really see your problem - why don't you simply replicate this diagram in SEM Builder? You don't tell us what is latent and what is observed in your model. In addition, whether you need multilevel latent variables is more of a substance question about what is the correct model than an estimation issue.