| Message/Author |
|
|
|
|
Hi there,
I'm using LCA and will use multi-traj model to find patterns in multi-tissue involvement in joint diseases. I have data for both knees for part of my dataset while for the other half I have data for only one knee. My question is if it is possible to account for the correlated observations in the analyses if I include the two knees of the same person as distinct variables. Thank you very much |
|
|
|
|
| You can account for correlated observations by letting the two processes be influenced by the same - or related - latent variables (latent class variables or growth factors). |
|
|
|
|
| Thank you for the response. Does it mean that I can include the correlated observations in the same model(GBTM)? Thanks again |
|
|
|
|
| Right. See e.g. Ug ex 6.13 for an example where the 2 outcomes are correlated via growth factors (an analogous approach is available using 2 latent class variables). |
|
|
|
|
| Ok, thanks very much for your help! |
|
|
| Back to top |
