No. The sample size needed depends on the characteristics of the variables and how separated the classes are. Only a Monte Carlo study can determine the necessary sample size for your situation.
Erika Wolf posted on Wednesday, September 01, 2010 - 3:58 pm
Can you point me towards a script for a Monte Carlo study for an LCA? I see reference in other posts to examples on the installation disc, but Mplus was delivered electronically, so I don't have a disc. I'm trying to figure out how to specify the percent of cases in each latent class in the population. Thanks.