Hi all,

I would like to analyze a two-period survey data from an impact evaluation. Dependent and independent variables are Likert-scale questions - I was thinking of using Generalized Structural Equations Modelling to estimate the latent variables captured using the Likert scale variables and see whether the treatment had an impact on the latent variables (essentially run a Difference-in-difference regression using a GSEM approach). I was wondering whether this would be possible - example 46g in the SEM documentation (Endogenous treatment effects) only compares unionized vs non-unionized populations in a single-period survey.

Thank you