Individually-varying times of observa... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Victor Heh posted on Thursday, October 15, 2009 - 9:09 am
I have a repeated measurement data collected in 4 years on 1000 participants. In year 1, data was collected on 592 cases that were followed through year 4. From year 2 to year 4, while the original cases were being followed, new cases were added each time.
There are 9 continuous nutritional indicators (Observed dependent variables)
All covariates are time-varying
Primary covariate of interest is (1=caregiver, 0= non-caregiver) which could change over time.

My questions are:
1. Can we use data from all 1000 cases to study growth?
2. If yes, what time points (1-4) should be assigned to later additions? For example, for cases that entered in year 4, do we assign time4 or time1? Is this really the same as Individually-varying times of observations?
3. Is year (year1, year2, year3, year4) the best metric for time? (Age data is available but I think has some measurement issues since these people are grandparents in a developing country.
4. Would it make sense to follow the “Marital Status change and Alcohol use model (Curran, Muthen, & Harford, 1998)?
Thank you!
 Linda K. Muthen posted on Friday, October 16, 2009 - 8:18 am
1. You can use the 1000 cases. This assumes they come from the same population. You have to be able to support this assumption.
2. The timepoint assigned depends on the time axis. If it is age, then a new individual should be added at the appropriate age.
3. The best metric for time depends on what you are measuring and what you expect to happen as the variable is repeatedly measured.
4. I don't think this is relevant.
 Victor Heh posted on Friday, October 16, 2009 - 9:14 am
Thank you! This is very useful.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message