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

  • GSEM: Using a latent variable as an explanatory variable in the multinomial logistic regression


    Is it possible to use CFA outputs, latent variables, as explanatory (independent) variables in a multinomial logistic regression model in a GSEM model? I thought this would be as easy as adding a path from the latent variable to my categorical dependent variable but it does not seem to work.

    So what I do is I add a measurement component first, which works fine by itself, then I add a regression component, which works fine by itself as well, then add a path from the latent variable to the dependent variable in the regression component. But, I guess, rather than treating the latent variable as an explanatory variable for the regression, it treats my dependent variable as another source variable for the latent variable.

    I hope this makes sense. I am normally an R user and an SEM newbie, so apologies if I am missing things that are obvious.


  • #2
    This would make more sense if you showed what code you're using, preferably in code delimiters (see my signature for details). I would guess that you have an arrow facing the wrong direction, but that's just a guess until we see code.
    Please use the code delimiters to show code and results - use the # button on the formatting toolbar, between the " (double quote) and <> buttons.

    Please use the command -dataex- to show a representative sample of data; it is installed already if you have Stata 14.2 or 15.1, else you can install it by typing

    ssc install dataex