|
Message/Author |
|
Judy Black posted on Monday, December 10, 2012 - 3:04 am
|
|
|
Dear Dr. Muthen, I would like to examine a moderation effect using change scores. There are six observed variables: a1,a2,b1,b2,y1,y2.I would use these observed variables to create latent variables: change a = a2-a1; change b = b2-b1; change y = y2-y1; Using these latent variables, I would like to examine the interaction effect between 'change a' and 'change b' on 'change y'. I could create 'change a','change b',and'change y' in the MODEL command. My question is, how can I create the interaction item between 'change a' and 'change b'? Thanks very much! Best, Judy |
|
|
Why are the difference scores latent? Aren't they the difference between two observed variables? |
|
Back to top |
|
|