Testing longitudinal structural invar... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Sami Yli-Piipari posted on Tuesday, December 11, 2012 - 1:31 pm
Drs, I am interested to test whether theorized relationship is invariant across two time points (T0 & T1). I would like to make sure that I use appropriate modeling approach to test the hypothesis. I have 3 latent variables (4 factors each). An assumption is that exercise climate relates to perception of autonomy and that, in turn, relates to exercise behavior. Specifically, my questions are:
1. Should I test measurement invariance where all variables are in one model: I.e. climate -> autonomy->behavior together with regressive paths between climate (T0 & T1), autonomy (T0 & T1), and behavior (T0 & T1). And then testing partial and full measurement invariance using normal procedures.
2. OR should I test, firstly, measurement variance separately between each variable i.e. climate (T0 & T1), autonomy (T0 & T1), and behavior (T0 & T1). And, secondly, adding the regressive paths after measurement variable is established.
My primary interest is whether theorethical model is invariant across time. Thank You very much.
 Linda K. Muthen posted on Wednesday, December 12, 2012 - 8:50 am
Number 2.
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: