Hi everyone,
I am using the LCA Distal Stata Functions plugin developed by Penn State (https://www.latentclassanalysis.com/...-distal-stata/) to use latent classes as a predictor for a distal outcome. In a next step, I would like to introduce third variables to check for confounding effects. Unfortunately, there does not seem to be an option within the plugin to introduce third variables, and I couldn't find a solution to this problem on this forum or elsewhere.
So I was wondering if it might be impossible to integrate third variables in computations with the plugin? Or am I maybe (hopefully) missing something?
I am using the Penn State plugin because I am new to latent class analysis and I find calculations are a lot easier with said plugin than with base Stata. Additionally, since I am using Stata 15, I believe I cannot use the full spectrum of latent class analysis computations available in newer Stata versions (I might be mistaken). Nonetheless, if there is a way to solve my problem with base Stata, this would obviously be preferable to not being able to solve it.
Any suggestions for possible solutions would be very much appreciated!
I am using the LCA Distal Stata Functions plugin developed by Penn State (https://www.latentclassanalysis.com/...-distal-stata/) to use latent classes as a predictor for a distal outcome. In a next step, I would like to introduce third variables to check for confounding effects. Unfortunately, there does not seem to be an option within the plugin to introduce third variables, and I couldn't find a solution to this problem on this forum or elsewhere.
So I was wondering if it might be impossible to integrate third variables in computations with the plugin? Or am I maybe (hopefully) missing something?
I am using the Penn State plugin because I am new to latent class analysis and I find calculations are a lot easier with said plugin than with base Stata. Additionally, since I am using Stata 15, I believe I cannot use the full spectrum of latent class analysis computations available in newer Stata versions (I might be mistaken). Nonetheless, if there is a way to solve my problem with base Stata, this would obviously be preferable to not being able to solve it.
Any suggestions for possible solutions would be very much appreciated!
