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

  • Ordinal variables in Latent profile analysis

    Dear Stata users,

    I have again a doubt on the latent profile analysis in Stata (gsem command). Among my independent variables, I have an ordinal categorical variables concerning income, with three levels (low, medium, high). I am not sure how to interpret the estimated coefficients of such variables when I compare this variables across classes. What I obtain are the cut-off points. Should I look at the magnitude or at the difference between the value of the cut-off points? I am sorry but I can’t export the results as I am working inside the system of a national statistics institute and it’s strictly forbidden. Thank you