Using summary data in multi-group ana... PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Han, Chang Hoon posted on Tuesday, August 28, 2018 - 6:31 pm
I want to do an alignment analysis.
However, I want to use input data as mean and covariance matrix rather than raw data.

I referenced EXAMPLE 13.2: MEANS AND A COVARIANCE MATRIX AS DATA in the Mplus guide ver.8.

In the case of multi-group analysis, input data for mean and a covariance matrix should be set for each comparative group.

Is it possible to set the mean and a covariance matrix input data separately for each group in the multi-group analysis?

And if this is possible, how do I implement it in mplus code?
I wish I could see an example.
 Linda K. Muthen posted on Wednesday, August 29, 2018 - 1:08 pm
See pages 539-40 of the current user's guide on the website.
 Han, Chang Hoon posted on Wednesday, August 29, 2018 - 8:10 pm
Thank you.
I should have checked more carefully.
I asked a simple and trivial question.
Nevertheless, thank you for your kindly reply.
 Han, Chang Hoon posted on Thursday, August 30, 2018 - 12:19 am
I analyzed the contents again by referring to the answers.
CFA was analyzed, but alignment could not analyzed.
I checked out file.

///////////////////////
!!!- output - from here
INPUT INSTRUCTIONS

TITLE: align

DATA:
FILE IS summary_dat.txt;
TYPE IS MEANS COVARIANCE;
NGROUPS = 2;
NOBSERVATIONS = 199 212;

VARIABLE:
NAMES= y1-y5;

ANALYSIS:
ESTIMATOR= ML;
alignment= fixed;
PROCESSORS= 2;

MODEL:
f by y1 y2 y3 y4 y5;

OUTPUT: align;

INPUT READING TERMINATED NORMALLY

align

!!!- output - END here
//////////////////////

The above output is done.
And the output screen does not popuped.

I'am using Mplus ver 7.4.

The below is my input data.
5 indicator, means & cov structure, 2group.


-1.779 -3.0 -3.77 -2.334 2.024
1.000
0.634 1.000
0.811 0.879 1.000
0.799 0.750 0.969 1.000
0.574 0.586 0.698 0.720 1.000
-1.053 -2.397 -2.592 -1.729 2.991
1.205
0.757 1.330
0.817 1.163 1.239
0.699 0.796 0.911 0.828
0.352 0.537 0.547 0.444 0.919
 Han, Chang Hoon posted on Thursday, August 30, 2018 - 12:35 am
I tried using mplus ver 8.1.
Same problem occurred.
 Linda K. Muthen posted on Thursday, August 30, 2018 - 8:49 am
Please send the output and your license number to support@statmodel.com.
 Youngshin Ju posted on Wednesday, March 04, 2020 - 8:48 am
To. Mplus team

Hello,

I'm trying to run a multi-group categorical CFA with summary data (tetrachoric correlation matrix) across 32 groups for 7 binary items. However, I saw the error below.

##################################

*** ERROR in DATA command
Multiple group analysis with categorical dependent variables requires
means as part of the summary data.

#################################

I don't understand this error because it's impossible to calculate the average in the binary items. What should I do?

Thanks a lot in advance!
 Bengt O. Muthen posted on Wednesday, March 04, 2020 - 11:36 am
Please send your output and data to Support along with your license number.
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: