Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • SEM: Latent variable as a predictor of observed variable

    Good day!

    Is it possible for a latent variable to become a determinant or predictor variable of an observed variable?
    I am trying to create a path diagram in Stata's SEM builder. My latent variable is "fiscal illusion" which is determined by some institutional variables such as government effectiveness, political stability, rule of law, etc. which are all observed variables. So I have 5 arrows pointing towards fiscal illusion.. Now I want to measure the effect of fiscal illusion on corruption, tax revenue, government expenditure and economic growth.. If I put arrows from the latent variable "fiscal illusion" towards observed variables corruption, tax revenue, government expenditure and economic growth, will it make the latent variable a predictor variable? Or will this make the observed variables "indicators" of the latent variable?

    Very much hoping for a reply. Thank you.

  • #2
    I believe your query is related to the concepts of "reflective" and "formative" models. I suggest you take a look at both key words under SEM.
    Best regards,

    Marcos

    Comment

    Working...
    X