Categorican predictors PreviousNext
Mplus Discussion > Structural Equation Modeling >
 ffan posted on Friday, February 21, 2014 - 6:43 pm
Hi Dr. Muthen,

I am using path analysis with Mplus to get direct and indirect effect of a number of predictors on math achievement using the TIMSS data. All the predictors are composite scores from Likert type data and they are categorical(but not in integers). I wanted to use MLM to estimate both the direct effect and indirect effect. But there are three problems:1)Estimator MLM is not available for analysis with categorical variables; 2)CATEGORICAL option is used for dependent variables only;and 3)MODEL INDIRECT is not allowed with TYPE=IMPUTATION.

So I wonder how I could handle the three problems. Thanks a lot.
 Bengt O. Muthen posted on Saturday, February 22, 2014 - 4:02 pm
I think sums of Likert scales can be treated as continuous variables especially if they play the role of covariates only in the model.
 ffan posted on Saturday, February 22, 2014 - 5:43 pm
Thanks Dr. Muthen. A further question, can Mplus estimate indirect effects if I used the command "TYPE=IMPUTATION"? If not, that's OK. I might just have to calculate by myself.
 Linda K. Muthen posted on Tuesday, February 25, 2014 - 1:18 pm
You can use MODEL CONSTRAINT with TYPE=IMPUTATION to compute indirect effects and obtain their standard errors.
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