Attrition analysis PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Joy Choi posted on Thursday, April 11, 2013 - 11:35 pm
I am running attrition analysis using sex, income, maternal depression with IV (attachment) and DV (social emotioanl functioning outcomes using growth curve midpoint intercept).

This is one of my syntax for this analysis, but results is weird and not presenting anything.

MODEL:

! INTERCEPT LOADINGS
I_bex BY bexk@1 bex1@1 bex2@1 bex3@1 bex4@1 bex5@1 bex6@1;
S_bex BY bexk@-3 bex1@-2 bex2@-1 bex3@0 bex4@1 bex5@2 bex6@3;

! REGRESSION
I_bex sex incom mdep on attchment;
I_bex sex incom mdep on other;

* bex: externalizing behavior problems
mdep: maternal depression at the child's age of 1 month
other: sort of missing data. Among entire sample without attachment data.

Questions are:
1. what is wrong with this syntax?
2. How do I compare "attachment" and "other"? chi-square differences or model fit?
 Linda K. Muthen posted on Friday, April 12, 2013 - 4:37 pm
1. The syntax is not a properly specified growth model. I would suggest using the | language which sets the model up for you and also has defaults appropriate for growth models:

i_bex S_bex | bexk@-3 bex1@-2 bex2@-1 bex3@0 bex4@1 bex5@2 bex6@3;

2. If attachment and other are both binary, you can compare the coefficients using MODEL TEST.
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: