Parallel Process and Multi-level Mode... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Rosemay Remigio-Baker posted on Thursday, July 28, 2011 - 5:13 pm

I am new to MPLUS and was interested in using parallel process latent growth curve analyses simultaneously with multi-level modeling (2-levels: time, individual and neighborhood). First, is this possible to do since I understand that the parallel process requires a wide-formatted data, whereas multi-level modeling requires a long-formatted data? If so, what would you recommend for a data structure? Also, is it correct to say that when looking at our WITHIN results, the estimates/SEs account for the correlation of individuals living within the same neighborhood? Any guidance you could provide would be much appreciated.

Thank you.
 Bengt O. Muthen posted on Thursday, July 28, 2011 - 5:28 pm
Yes, this can be done. You have as columns in your data the two processes for all time points, say:

y1, y2, ..., yT, z1, z2, ..., zT plus covariates

and repeat that over all people in all clusters (data don't have to be given cluster by cluster).

And, yes, the SEs then account for the intra-neighborhood correlations because you model this correlation by allowing Between (Level2) level variances.
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