Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: SEM: Using individually-varying time scores in latent growth model


From   Cameron McIntosh <cnm100@hotmail.com>
To   STATA LIST <statalist@hsphsun2.harvard.edu>
Subject   RE: st: SEM: Using individually-varying time scores in latent growth model
Date   Fri, 21 Oct 2011 21:19:05 -0400

Jacqui,
In Mplus you should have been using the TSCORES option -- I'm not sure exactly how you implemented the 'time in study' approach you mention, or what methodological reference you found it in. Anyway, I don't think there is an automated procedure in Stata's SEM to do this. You should have a look at:
Blozis, S.A., & Cho, Y.I. (2008). Coding and centering of time in latent curve models in the presence of interindividual time heterogeneity. Structural Equation Modeling, 15(3), 413-433.
King, D.W., King, L.A., McArdle, J.J., Grimm, K., Jones, R.T., & Ollendick, T.H. (2006). Characterizing Time in Longitudinal Trauma Research. Journal of Traumatic Stress, 19(2), 205–215.http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2151057/pdf/nihms9820.pdf ;
You may need gllamm or xtmixed to implement the methods.
Rabe-Hesketh, S. & Skrondal, A. (2008). Multilevel and Longitudinal Modeling Using Stata (2nd ed.).  College Station, TX: Stata Press. 
Cam
----------------------------------------
> Date: Fri, 21 Oct 2011 14:06:05 -0700
> From: jhomel@uvic.ca
> To: statalist@hsphsun2.harvard.edu
> Subject: st: SEM: Using individually-varying time scores in latent growth model
>
> Dear List,
>
> I am using SEM (Stata 12) to estimate a latent growth model of
> depression measured at four time points between adolescence and early
> adulthood. Previously I have estimated this model in MPlus, and now I am
> trying to learn how to produce the same results using SEM. The below
> model works fine-
>
> sem (dep_3 <- Idep@1 Sdep@0 _cons@0)
> (dep_5 <- Idep@1 Sdep@1 _cons@0)
> (dep_7 <- Idep@1 Sdep@2 _cons@0)
> (dep_9 <- Idep@1 Sdep@3 _cons@0),
> latent(Idep Sdep)
> var(e.dep_3@var e.dep_5@var
> e.dep_7@var e.dep_9@var)
> noconstant means(Idep Sdep) method(mlmv)
>
> However, although observation points were two years apart, not all
> participants were interviewed at the same time, leading to
> individually-varying times of observation. Earlier analyses using MPlus
> used 'time in study' to represent time, rather than 0, 1, 2, 3. Is there
> a way to incorporate these varying time scores into a growth model using
> SEM?
>
> Kind regards,
>
> Jacqui
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index