Define - standardize with 2nd var's m... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Andrew Bender posted on Tuesday, July 09, 2013 - 7:40 am
Drs. Muthen,

I'm running a latent difference score model in Mplus 7 for comparison of two measurement occasions (e.g., Time1, Time 2). Moreover, as we build the model, there will be multiple different variables for Time1/Time2 (eg., x_T1, x_T2, y_T1, y_T2, etc.) for which the variances may vary considerably. This suggests that standardization may be desirable or helpful.

I have been told that if I wish to standardize the data, that standardizing both variables using their respective means and standard deviations is not acceptable as this eliminates the change in means and variances that we are trying to model. Thus, the recommended approach is to use the Time1 means and standard deviations to standardize the Time2 data.

I am trying to use the DEFINE command to produce a z-score formula [(T2-T1mean)/T1stddev]. However, I am not clear on the best way to run this define command, to include the mean for Time1.

Any assistance you might provide is greatly appreciated.

thanks in advance
 Linda K. Muthen posted on Wednesday, July 10, 2013 - 12:31 pm
I would not standardize variables. If you have a variable with a large variance, I would divide it by a constant.
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: