Convert LISREL syntax to MPlus: a que...
Message/Author
 Shan Yeung posted on Tuesday, December 10, 2013 - 11:48 pm
Dear Prof. Linda K. Muthen and Bengt Muthen,

I am converting a LISREL syntax to a Mplus syntax, not feeling that sure of my changes. Would you please have a look at these changes and give me some comments.

Thank you so much for your time and help.

Grace

LISREL syntax

observed variables x1 x2 x3 x4 x5 x6 x7 x8 x9 x10

Latent Variables: A C M P G

x1 x2 x3 = A
x4 x5 x6 = C
x7 x8 x9 = M
x10 = 1 * P
set the error variance of x10 to 0

let the errors between x10 x7 correlate
let the errors between x6 x5 correlate
let the errors between x10 x8 correlate

A C M P = G

number of Decimals = 2

Print Residuals
Method of Estimation: Maximum Likelihood
path diagram
End of Problem

Mplus syntax by Grace

variable: names = x1 x2 x3 x4 x5 x6 x7 x8 x9 x10;

missing = all (-99);
usevar = x1 x2 x3 x4 x5 x6 x7 x8 x9 x10;

analysis:
type = general;
ESTIMATOR = MLR;

model: A by x1 x2 x3;
C by x4 x5 x6;
M by x7 x8 x9;

! fixed the path from x10 to P to 1 and fixed the error variance to 0
P by x10;
x10 @ 0;
x10 with x7;
x6 with x5;
x10 with cjom2;
G by A C M P;
output: sampstat standardized stdyx
 Linda K. Muthen posted on Wednesday, December 11, 2013 - 6:59 am
Run the model you have specified in Mplus. If you don't get the same results as in LISREL, send the Mplus output, the LISREL output as a pdf or doc file, and your license number to support@statmodel.com.