LCA comparing groups PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Lauren Armstrong posted on Friday, May 25, 2018 - 6:12 am
I have done a LCA where I have 3 classes. All variables are continuous. I have a couple of questions:

1) I want to compare each of the classes on levels of self-esteem. I have run the syntax:

auxiliary = SE(BCH);
type = mixture;

The output gives me a chi-square test for the differences between classes, however I don't think this is appropriate due to the scores being continuous. How can I program this to give me t-tests or an ANOVA?

2) How can I produce descriptive statistics for other measures I may want to include for each of the latent classes?

Many thanks!
 Bengt O. Muthen posted on Friday, May 25, 2018 - 12:43 pm
1) Chi-square testing is also for continuous outcomes - often called Wald testing.

2) I don't know what you mean.
 Lauren Armstrong posted on Monday, June 04, 2018 - 1:10 am
1) thanks!

2) Sorry for not being clear. I have included measures of depression & anxiety as well as questionnaires that capture demographic information such as age and gender in my dataset. How do I see, for each of my 3 latent classes, what their levels of depression are and what their mean age is etc?
 Bengt O. Muthen posted on Monday, June 04, 2018 - 12:56 pm
Use BCH or DCAT in line with Mplus Web note 21 (see also table at the end).
 Lauren Armstrong posted on Wednesday, June 06, 2018 - 8:52 am
Thank you!
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message