Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: Multilevel SEM

From   Christopher Salas-Wright <>
Subject   st: Multilevel SEM
Date   Wed, 21 Dec 2011 10:05:26 -0500

I have been having difficulty using multilevel (or second-order) structural
equation models using Stata/IC 12.0.  I have had no problem running
firstorder structural equation models with a combination of observed
exogenousvariables and latent endogenous variables using, for example,
the followingsyntax:
sem (friend5 friend6 <- PEER) (gras7 gras8 <- BELIEFS) (drugcut
drugcritdrugguil alc1 ciga1 drug1 <- SUBST)  (rcope_d -> PEER)
(hodge5_d-> PEER)(empathy_d -> PEER) (ptsd_d -> PEER) (goodlivefam ->
PEER) (edusingl->PEER) (gender-> PEER) (age1-> PEER) (PEER -> BELIEFS)
(BELIEFS -> SUBST)(hodge5_d-> SUBST) (gender->SUBST)
(goodlivefam->SUBST) if _mi_m ==0, cov(e.drugguil*e.drugcut) cov
(e.drugcrit*e.drugcut)cov(e.drugcrit*e.drugguil) method (mlmv) stand
However, I receive a long error message (see below) every time that I
tryto write the syntax for a second order model for my dependent
variable(SUBST), using the following syntax:
sem (friend5 friend6 <- PEER) (gras7 gras8 <- BELIEFS) (drugcut
drugcritdrugguil <- ABU) (alc1 ciga1 drug1 <- USE) (ABU USE <- SUBST)
(rcope_d ->PEER) (hodge5_d-> PEER) (empathy_d -> PEER) (ptsd_d ->
PEER) (goodlivefam-> PEER) (edusingl-> PEER) (gender-> PEER) (age1->
PEER) (PEER -> BELIEFS)(BELIEFS -> SUBST)   if _mi_m ==0, method
(mlmv) stand
This is the error message:
"Fitting baseline model:
Iteration 0:   log likelihood = -3507.3904Iteration 1:   log
likelihood = -3504.3428Iteration 2:   log likelihood =
-3504.3299Iteration 3:   log likelihood = -3504.3299
              _sem_iv():  3301  subscript invalid
_sem_build__params():     -  function returned error
_sem_build():     -  function returned error           _sem_parse():
  -  function returned errorst_sem_parse_and_build():     -  function
returned error                <istmt>:     -  function returned

Is anyone aware of the mistake I might be making?  Thank you!  My
samplesize is 300.

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index