To expand on the issue I'm having a bit:

I am trying to build a latent class model. The dataset I am working with contains some scales related to shopping habits, plus several demographic items (age, race, income, etc.).

When I try to build the model with all variables added, it cannot converge. I am able to get convergence with a model (5 classes) that just includes the shopping habit scales, but I would still like to see the demographic differences between each class.

Would it be OK for me to take the classes I have developed and use them as predictors in a series of chi-square analyses or ANOVAs or something to examine how each of the 5 classes differ on race, income, age, etc? Or would this not be methodologically sound?

Alternatively, does anyone have any general advice on how to deal with a model when you can't seem to get it to converge? Thanks!


1 Answer 1


This is a very general reply, but typically it would be good to have theory-based idea on what variables you want to put in cluster analysis instead of deciding that on the basis of whether the model works or some other post-analysis statistical criterion. But in my understanding it's fairly common to form the classes (i.e. run the LCA) based on some somewhat coherent group of variables (such as shopping habits) and then investigate cluster membership's relations with other variables (such as demographics) in a further analysis (e.g. chi-square tests or some form of regression). Whether this is a good idea is a very broad question that requires subject knowledge and knowledge about the particular research questions and goals of the study.

  • $\begingroup$ Thank you for the reply! Do you happen to know of any papers or other sources that use that kind of methodology? (i.e., forming a model based on a set of coherent variables and then investigating relationships among the clusters and demographic variables in follow-up analyses)? $\endgroup$ Commented Jun 16, 2023 at 23:30
  • 1
    $\begingroup$ I can't think of a particular paper right now as it's been a few years since I used LCA but you can probably easily find several by searching e.g. google scholar. Here is a methodological paper in which the authors describe this practice. $\endgroup$
    – Sointu
    Commented Jun 17, 2023 at 8:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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