Ordinal model PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Scott C. Roesch posted on Tuesday, January 26, 2010 - 12:23 pm
I have an outcome with 3 ordered categories coded the following: (0=unlikely, 1=somewhat likely, 2=very likely). How does MPlus code this variable with respect to comparing cumulative logits? Is the first comparison between the unlikely vs. somewhat likely and very likely groups; and the second comparison between the unlikely and somewhat likely group vs. very likely)? Thanks!
 Linda K. Muthen posted on Wednesday, January 27, 2010 - 10:21 am
That is not correct. Please see slides 49-57 of the Topic 2 course handout for a full description.
 Scott C. Roesch posted on Sunday, January 31, 2010 - 12:46 pm
Thanks Linda, the slides helped. One last question, are the estimates provided by MPlus for nonlinear models (e.g., logistic, ordinal) population-average or unit-(subject-)specific. Thanks again!
 Linda K. Muthen posted on Sunday, January 31, 2010 - 1:19 pm
I don't think these concepts apply to logistic regression.
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: