Strange Error Message with ESEM PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
Message/Author
 Samuel S Nordberg posted on Tuesday, May 24, 2011 - 7:21 am
Dear Linda and Bengt,

This error message pops up when running the script below. I believe that the program runs into an infinite loop, as these two errors are repeated over and over, until the program exits (after about 2 seconds)?

*** ERROR
Variable names must begin with an alphabet character:
*1
*** ERROR
Label is not a valid parameter label. This label will not be
available for use in MODEL CONSTRAINT: *1

For the life of me, I cannot figure out what the problem is with the code, however. Here is the actual code:

VARIABLE: NAMES ARE y1-y107;
CATEGORICAL ARE y1-y107;
ANALYSIS: ESTIMATOR = WLSMV;
MODEL:
f1-f2 by y1-y37(*1);
f3 by y62-y76;
f4 by y38-y61;
f5 by y89-y102;
f6 by y77-y88;
f7 by y103-y107;
f3 ON f1-f2;
f4 ON f1-f2;
f5 ON f1-f2;
f6 ON f1-f2;

Thank you in advance for your help and advice.
 Linda K. Muthen posted on Tuesday, May 24, 2011 - 7:36 am
Please send the full output and your license number to support@statmodel.com.
 Fazil Aliev posted on Thursday, June 23, 2011 - 10:51 am
Dear Linda and Bengt,
I'm trying to compare Exploratory Factor analysis with direct method and with exploratory structural equation modeling (ESEM)for the same data with same rotation(GEOMIN) and estimator(MLR) options. Correlation coefficients between factors are similar but different.
For EFA corr(F1,F2)=0.524, corr(F1,F3)=0.189 and corr(F2,F3)=0.192
For ESEM corr(F1,F2)=0.643, corr(F1,F3)=0.254 and corr(F2,F3)=0.294
What can cause this difference? All other model parameters and variable descriptives look like same.
Thanks
 Linda K. Muthen posted on Thursday, June 23, 2011 - 11:06 am
ESEM and EFA with the same rotation and the same estimator should yield the same results. Make sure the estimators are the same. If this doesn't solve it, send the outputs and your license number to support@statmodel.com.
 Thomas Schlenzig posted on Sunday, January 26, 2014 - 10:48 pm
Dear Linda and Bengt,

Iím doing ESEM at the stage of testing equality of factor variances and factor covariances. Using the model that it is described in the Mplus guide section 5.27 i got the fatal errors 1021. Like you suggested in 2004, skipping equality of factor covariances results in error 1001. Can you help me with this errors. Skipping the covariances is not really an option because it does not allow to go on with ESEM in order to better understand my model.

VARIABLE: NAMES ARE r1-r9 c1-c12 f1-f8 case;
USEVAR ARE r1-r9 c1-c12 f1-f8;
Grouping is case (1=auto 2=energy);

Analysis: iterations = 500000;
rotation = oblimin;

MODEL: rel cog fun by r1-r9 c1-c12 f1-f8 (*1);
rel with cog (1);
rel with fun (1);
cog with fun (1);
rel-fun@1;

Output: MODINDICES (all) TECH1;


Thankís a lot and best regards,
Thomas
 Linda K. Muthen posted on Monday, January 27, 2014 - 8:38 am
Please send the output and your license number to support@statmodel.com.
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: