Message/Author 

Joel Nigg posted on Saturday, January 21, 2012  4:38 pm



Is there an easy way to get a correlation matrix from MPLUS (either as output to output file, or as a matrix data file it generates), without running a factor or other complex model analysis first. The reason I want this is to see correlations with missing data handled by FIML Thank you Joel 


Yes. Do a TYPE = BASIC; with no MODEL command and the SAMPLE option of the SAVEDATA command. 

Joel Nigg posted on Sunday, January 22, 2012  5:46 pm



Linda, Got it. I see it in the manual now; just didn't know to look there. Thanks much! 


Hi, I have only just begun using MPlus. I loaded the dataset and asked for Type=basic; hoping to getting means and std deviations. But the output only gave variances and no other descriptive statistics. What should I do to get the means, std deviations as well as minimum and maximum values for the variables listed after Usevariables ? 


You should get means followed by covariances and then a set of other descriptive statistics. We give variances rather than standard deviations. The square root of a variance is a standard deviation. You may be using an older version of the program. 


I am using MPlus 7 and still seem unable to generate means for variables; only variances are shown. Is there anything else I could do to troubleshoot? 


Decsriptive statistics give means, variances, and covariances. You would need to send the output and your license number to support@statmodel.com for me to understand why you don't get means. 


I seem to have figured it out by accident. There was a warning being thrown that the variance of one of my variables (ID variable) was out of range, i.e. >100,000. After I specifically identified it as the ID variable MPlus has stopped throwing that error/warning and also is giving all the other descriptive statistics for all the other variables. I did not realize that if MPlus detected an error on one variable it would stop processing all the other steps for the remaining variables as well. That is good to know. Thank you Dr. Muthen for responding to my queries. 


ok. 


Dr. Muthen, Does it mean that I should manually calculate the square root of a variance to know my standard deviations? (I am using Mplus 8.1, and I ran a type=basic) 


I don't know what you refer to. 

Back to top 