|
|
Bootstrap CI for comparing classes |
|
Message/Author |
|
|
Usually, when I run an LCA, the output contains a section titled, "Latent Class 1 Compared to Latent Class 2" which gives me p-values for comparisons of the item distributions between classes 1 and 2. However, when I use the BOOTSTRAP and CINTERVAL(BOOTSTRAP) options (which require me to change the estimation method to ML), that section is no longer there. Why not? Is there some way to get bootstrap confidence intervals for the odds ratio to compare item distributions between classes? |
|
|
This does not make sense to me. Please send the full output and your license number to support@statmodel.com. |
|
Back to top |
|
|