Nominal Outcome in Multilevel-Analysis PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Daniel Schwarzkopf posted on Thursday, February 04, 2010 - 1:35 am
Dear Dr. Muthen,
in a Multilevel-model I want to predict membership in one of three groups on the within level. I use a random slope in predicting group membership.
On the between-level I want to predict the random slope from a between level variable.
In multinomial Regression typically there are two slopes for nominal outcome with three values. But the output only shows the results of predicting one random slope on the between level. Is there a way to get both slopes? Or does it just dont work this way?

Thank you.
 Linda K. Muthen posted on Thursday, February 04, 2010 - 7:43 am
If you are specifying a multinomial logistic regression, you will obtain two slopes if you have three categories. I would have to see your full output and license number at support@statmodel.com to understand exactly what you are doing.
 Hoda Vaziri posted on Thursday, July 31, 2014 - 8:31 am
I want to do a multinomial logistic regression and my data is nested. but I'm not sure if I should do a multilevel analysis. Is there anything like ICC for nominal variables in Mplus?
Thanks.
 Linda K. Muthen posted on Thursday, July 31, 2014 - 9:51 am
No, there is not. Run the model with TYPE=COMPLEX and TYPE=GENERAL. If you see differences in the standard errors, this would indicate nesting should be taken into account.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: