|
|
Different specifications for LCA usin... |
|
Message/Author |
|
Anonymous posted on Monday, June 25, 2012 - 8:01 am
|
|
|
I am trying to find examples of how to specify in Mplus the following 3 step approaches to using covariates in LCA: most likely class regression, probability regression and probability-weighted regression. For example, supposing I have 4 latent classes and wish to explore the relationship between class and age. I have found the following example based on an initial 4 class run: Variable: names are age cprob1 cprob2 cprob3 cprob4 class; usevariables are age cprob1 cprob2 cprob3 cprob4; CLASSES = C (4); TRAINING= cprob1-cprob4 (PROBABILITIES); ANALYSIS: TYPE = MIXTURE; STARTS 100 10; MODEL: %OVERALL% c#1 ON age; c#2 ON age; c#3 ON age; Which type would this be an example of? Many thanks in advance on your helpful comments. |
|
|
This is probability regression. See the following paper which is available on the website for further information: Clark, S. & Muthén, B. (2009). Relating latent class analysis results to variables not included in the analysis. |
|
Anonymous posted on Tuesday, June 26, 2012 - 4:08 am
|
|
|
Many thanks. |
|
Back to top |
|
|