Multilevel LCA PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Jan Olav Christensen posted on Wednesday, October 28, 2015 - 2:02 pm
Dear Drs Muthen,

I am estimating a model with "type = twolevel mixture", with between- and within-level latent categorical factors and ordered categorical latent class indicators. The output does not give results in probability scale for latent class indicators. Is there a way to obtain that information, I suspect the thresholds cannot be converted to probabilities in the usual way? It is important that the latent classes are readily interpretable to non-statisticians.
 Bengt O. Muthen posted on Wednesday, October 28, 2015 - 3:01 pm
We need to see the output to say. Send to Support along with license number.
 Jan Olav Christensen posted on Thursday, October 29, 2015 - 12:15 am
Thank you for your reply, I apologize for not doing that in the first instance. I was under the impression that the lack of probability conversions for thresholds in a MLLCA was standard output. I thought maybe it could have to do with numerical integration.
 Miatta Echetebu posted on Thursday, February 04, 2016 - 5:40 pm
How do I determine if I have enough power to conduct a multilevel LCA? My data is from staff across different service sectors. We have a total of 8 sectors and 486 staff members. Is it reasonable to conduct a multilevel LCA under these circumstances?
Thank you.
 Bengt O. Muthen posted on Friday, February 05, 2016 - 6:07 pm
8 clusters is most likely too small. I would guess you need at least 20.
 Miatta Echetebu posted on Sunday, February 14, 2016 - 4:48 pm
Ok, 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