Poisson Regression with Mediation PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Audrey Clubb posted on Monday, March 03, 2014 - 9:08 am
Drs. Muthen,

Please forgive the elementary question, but I am trying to determine the correct syntax for a mediation analysis with a non-normally distributed outcome variable. In particular, I am struggling with significance tests for the indirect effects. The following is my present attempt. Could you help with clarifying the proper syntax for this analysis? Thank you.

Title: Mediation with Poisson;
Data: File is “c:\file.dat;
Variable: Names are DV MED P1 P2 P3 CLUST;
MISSING are all (-999);
CLUSTER are CLUST;
Usevariables are DV MED P1 P2 P3;
Count are DV (nb);
Analysis:
Type is COMPLEX;
Model:
DV on P1 (a1) P2 (a2) P3 (a3) MED;
MED on P1 (a1) P2 (a2) P3 (a3);
Model Test:
A1 = 0;
A2 = 0;
A3 = 0;
Model indirect:
DV IND MED P1 P2 P3;
Output: samp stand cinterval;
 Bengt O. Muthen posted on Monday, March 03, 2014 - 12:06 pm
You cannot have several labels on the same line without using semi colons.
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: