Descriptives for clusters
Message/Author
 EvavdW posted on Wednesday, November 01, 2017 - 8:55 am
Hi, I am working on a multilevel regression model with gender as predictor and several outcome measures.

Now, I would like to report the means and SD's for these outcome measures for males and females separately.

Is there an easy way to get this in Mplus?

I read about the cluster_mean option, but could not quite work it out.

Below is what I have currently, with Cito1 - Cito 5 being my outcome measures:

VARIABLE: NAMES ARE PN Gender Class1 Cito1 Cito2 Cito3 Cito4
Cito5;
USEV ARE Gender Class1 Cito1 Cito2 Cito3 Cito4 Cito5;
CLUSTER = Class1;
CLASSES = g(2);
knownclass = g(Gender = 0 1);
MISSING ARE ALL (999);
ANALYSIS: TYPE = mixture;

DEFINE: clusmean = CLUSTER_MEAN (Cito1 Cito2 Cito3 Cito4 Cito5);

With kind regards, Eva
 Linda K. Muthen posted on Wednesday, November 01, 2017 - 9:02 am
You can do a TYPE=BASIC using the GROUPING option with gender as the grouping variable and obtain sample statistics for males and females separately.
 EvavdW posted on Monday, November 06, 2017 - 8:00 am
Dear Linda, Thank you so much for the quick reply.

I've tried:

VARIABLE: NAMES ARE PN Gender x y;
USEV ARE Gender x y;
!CLUSTER = Class1;
CLASSES = g(2);
knownclass = g(Gender = 0 1);
MISSING ARE ALL (999);
ANALYSIS: TYPE = BASIC MIXTURE;

%overall%
x;
y;

%g#1%
x;
y;

%g#2%
x;
y;

But got the following error:
*** FATAL ERROR

CLASS 1 HAS 0 OBSERVATIONS.

When I look in my data file there are definitely participants with a score of 0. What may be the cause of the problem here?

With kind regards, Eva
 Linda K. Muthen posted on Monday, November 06, 2017 - 11:30 am
Please send the files and your license number to support@statmodel.com.