BSEM Measurement Invariance PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Shanshan Wang posted on Saturday, January 10, 2015 - 8:43 pm
Hi,

I have two questions.

Q1: In Muthen & Asparouhov's (2013) paper BSEM Measurement Invariance Analyis, there was an application with the PISA 2003 data in section 5. Is that script available?

Q2: This is the code I modified from Example 5.33 for binary data. Is it correct?

VARIABLE:
NAMES =
id group y1-y5;
USEVARIABLES =
group y1-y5;
CATEGORICAL =
y1-y5;
CLASSES =
c(100);
KNOWNCLASS =
c(group = 1-100);
ANALYSIS:
TYPE = MIXTURE;
ESTIMATOR = BAYES;
PROCESSORS = 2;
MODEL = ALLFREE;
MODEL:
%OVERALL%
f BY y1* y2-y5(lam#_1-lam#_5);
%c#100%
f@1;
[f@0];
MODEL PRIORS:
DO(1,5) DIFF(lam1_#-lam5_#)~N(0,0.01);
OUTPUT:
TECH1 TECH8;
 Bengt O. Muthen posted on Monday, January 12, 2015 - 11:06 am
Here is how we did it, including the thresholds:

ANALYSIS:
type = mixture;
estimator=bayes;
proc=2;
model = allfree;

model:
%overall%
f by y1-y8* (lam#_1-lam#_8);
[y1$1-y8$1] (tau#_1-tau#_8);

%c#40%
f@1;

model priors:
do(1,8) diff(tau1_#-tau40_#)~N(0,0.10);
do(1,8) diff(lam1_#-lam40_#)~N(0,0.01);
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: