A colleague of mine has asked me for help. She has a large amount of patient data involving clinical psychological measures (e.g., questions related to specific symptoms) gathered both before and after a psychological intervention in the same subjects. She believes there to be latent states within these data – representing different phases of the disorder (during the initial or final time point) or recovery (during the final time point). She is interested in applying a statistical model capable of taking data from 9 observed variables (representing answers to questions on a 4-point ordinal scale), each measured in ~2000 participants, and estimating the latent state for each participant at time one and at time two, and further estimating the transition probabilities between states from pre-measurement to post-measurement.


Being relatively new to this form of analysis, I am unsure how best to proceed. This sounds to me like some variant of a latent transition analysis, but I have been unable to find any examples showing how such an analysis could be implemented in any free analysis framework (e.g., R, Python). I would be most appreciative if someone could point me in the right direction.


Your colleague can apply a latent class analysis through which a latent variable relying on the clinical psychological measures can be identified. You may refer to the poLCA package by Lenzer and Lewis in R:


For the longitudinal analysis between point 1 and point 2, you may refer to the Latent Class Growth Analysis

Extensive details are provided by Feldman et al.:


In R, you may refer to this relatively new package by Proust-Lima et al.:



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.