How to weight indicator variables PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Kelcie Ralph posted on Friday, January 09, 2015 - 9:12 am
I apologize if this has been answered elsewhere.

I have a mixture model with 7 variables: three from a single day of data and four from an entire year. I have successfully identified five classes where each of the seven indicator variables were weighted equally.

However, a reviewer noted that I should weight the long-term data more heavily than short-term data. Is there a way to do that in Mplus? In other words, can I make some of the indicator variables "count more" in the model?

To be clear, I am NOT trying to use a complex survey design or sample weights.

Thank you.
 Bengt O. Muthen posted on Friday, January 09, 2015 - 4:56 pm
I don't see how you can do that. But perhaps you can do a confirmatory LCA with 2 latent class variables - one for the 3 items and one for the 4 items - and see how highly those 2 latent class variables correlate; if highly correlated, the question is moot.
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