Multilevel LCA, with 10 or more clusters PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Danielle Ompad posted on Saturday, April 19, 2014 - 4:06 pm
Hello!

I am trying to conduct a multilevel LCA where people are nested within 10 or more cities. I am identifying the city variable as nominal. I keep getting the following error:

*** ERROR in VARIABLE command
Unknown variable in NOMINAL option:
city

Here are the variable, analysis, and model parts of the program:

VARIABLE:
NAMES ARE id syeptot (15 variables) coneptot
(15 more variables) infeptot (15 more variabls) cnseptot
(14 final variables notests ptime city testres ;

Missing are . ;

USEVARIABLES ARE syeptot coneptot infeptot cnseptot city;
NOMINAL = city;

CLASSES = c(2);
CLUSTER = city;

ANALYSIS:
TYPE IS MIXTURE COMPLEX;
ALGORITHM=INTEGRATION;

STARTS=100;
LOGHIGH = +15;
LOGLOW = -15;
UCELLSIZE = 0.01;
ESTIMATOR IS MLR;
LOGCRITERION = 0.0000001;
ITERATIONS = 1000;
CONVERGENCE = 0.000001;
MITERATIONS = 500;
MCONVERGENCE = 0.000001;
MIXC = CONVERGENCE;
MCCONVERGENCE = 0.000001;
MIXU = CONVERGENCE;
MUCONVERGENCE = 0.000001;

MODEL:
%OVERALL%
syeptot WITH coneptot-cnseptot;
coneptot WITH infeptot cnseptot;
infeptot WITH cnseptot;

Could you please advise?

Many thanks,

Danie
 Bengt O. Muthen posted on Saturday, April 19, 2014 - 4:45 pm
Try taking your city variable off the Nominal list. The Nominal and other variable declarations are only for dependent variables used in the Model command.
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: