Coding of a predictor for growth curv... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Jungmeen Kim posted on Monday, May 08, 2006 - 3:14 pm
Hi Linda and Bengt,

I am trying to run a growth curve model with 7 times points of the outcome variables and test the effects of child maltreatment on growth factors. One of the predictors is onset of the maltreatment that I coded 1=infancy, 2=toddlerhood, 3=preschool period, 4=school age period, and 5=later school age period. In the sample, half of children were maltreated and the other half were non-maltreated (control group). If I want to include all children in the model, how should I treat the non-maltreated children for the onset variable? If I assign 0=no maltreatment, it doesn't make sense because 0 would indicate an earlier onset than 1 (infancy). Should I use a set of dummy variables using the non-maltreated group as a reference group? Thanks very much for your help!
 Linda K. Muthen posted on Monday, May 08, 2006 - 4:10 pm
I think you should have a set of five dummy variables.
 Jungmeen Kim posted on Monday, May 08, 2006 - 7:16 pm
Will try it. Thanks Linda!
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: