Dichotomous and ordinal predictors PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Max Nachbauer posted on Wednesday, May 04, 2016 - 8:15 am
Dear Mr. and Mrs. Muthén

I am running a multilevel model, more precisely an intercept-and-slope-as-outcome-model (TYPE = TWOLEVEL RANDOM). The predictors on level two are treated as manifest variables. Some of the predictors are continuous, some are ordinal, some are dichotomous. For the ordinal and dichotomous predictors, I am not sure, if I specifiy the model correctly.

Ordinal predictors: Can I integrate these predictors just normally OR do I have to specify the ordinal scale somehow OR do I have to create dummy variables?

Dichotomous predictors: Can I integrate these predictors just normally OR do I have to specify the dichotomous scale somehow?

Thank you in advance!
 Linda K. Muthen posted on Wednesday, May 04, 2016 - 9:00 am
In regression, predictors can be binary or ordinal. In both cases, they are treated as continuous. You don't need to do anything about their scales.
 Geilson Lima Santana Junior posted on Wednesday, February 26, 2020 - 12:29 pm
Dear Mr. and Mrs. Muthén: I have a similar question. I am running a mediation model with an ordinal predictor, two continuous mediators and one binary outcome. In this case, is the ordinal predictor also treated as continuous or do I have to create dummy variables?
 Bengt O. Muthen posted on Wednesday, February 26, 2020 - 4:38 pm
You should create dummies.
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: