|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Tests of significance for LCA indicators |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
Isaac posted on Thursday, June 07, 2012 - 8:00 am
|
|
|
Hi, I'm conducting an LCA with 10 categorical predictors (presence vs. absence of different psychiatric disorders in a large population). I've uncovered three classes that seem to provide the best model fit. However, I want to assess if the probabilities for any given disorder between classes is significantly different. For example, Class 1 demonstrates a 60% probability of Disorder A while Class 2 demonstrates a 30% probability of Disorder A. Can I determine that these probabilities are significantly different from each other? Thanks a lot |
|
Back to top |
|
|