Indirect effect - categorical ind & o... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 rs posted on Saturday, March 10, 2012 - 9:07 pm
Dear Dr Muthen,

I am trying to run an analysis to determine whether m mediates the effect of the following X variables on Y.

I have
-3 independent categorical variables i.e. level of education (3 categories), level on income( 3 categories) gender(2 categories) and age (continuous)
- a mediator which is a continuous variable (m) and
- one outcome variable which is an ordered categorical outcome (y1 -y3).

I have looked at the examples in the Mplus textbook however I am still unclear whether I can use model indirect when I have more than one categorical independent variable and if so, do I need to create a dummy variable.

I tried with the following syntax (which failed to run)


Model:
m on edu;
m on inc;
m on gender;
m on age;
y#1 y#2 on m;
yk#1 y#2 on edu;
y#1 y#2 on inc;
y#1 yk#2 on gender;
y#1 y#2 on age;

Model indirect:
Y1 y2 ind m edu;
Y1 y2 ind m age;
Y1 y2 ind m gender;
Y1 y2 ind m inc;
Y1 y2 via m edu age gender inc;

Analysis:
ESTIMATOR = MLR;

Would really appreciate if you could help me with some suggestions how to run this model.

Thank you.
 Linda K. Muthen posted on Sunday, March 11, 2012 - 11:33 am
You say you have an ordered categorical outcomes. It should be on the CATEGORICAL list. It seems you have it on the NOMINAL list. It should be referred to as

y1 on m;
y1 on edu;
y1on inc;
y1 on gender;
y1 on age;
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: