Reconciling Sample Size Counts PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Anonymous posted on Thursday, September 01, 2005 - 11:23 am
I am running a multigroup MIMIC analysis with Type = Missing and am examining the frequency of the data patterns generated by the Patterns output option. For all four of my groups,when I add up the frequencies of the data patterns, I get a number that is larger than the number of observations sample size)listed in the SUMMARY OF ANALYSIS output for the group. Could you please explain why this difference occurs?

Here are the details:

709 obs omitted bacause of missing value on grouping variable

Group1: sum of pattern freqs = 546 but sample size listed in output = 516

Group2: sum of pattern freqs = 594 but sample size listed in output = 555

Group3: sum of pattern freqs = 789 but sample size listed in output = 744

Group 4: sum of pattern freqs = 628 but sample size listed in output = 560.

I note that when I add the printed sample sizes and the 709 omitted observations, I get 3,084 which is the real number of observation in my data file.

516 + 555 + 744 + 560 + 709 = 3,084 (yea!)

Thank you very much for your assistance.
 Linda K. Muthen posted on Thursday, September 01, 2005 - 2:19 pm
You would need to send the input, output, data, and your license number to support@statmodel.com. This is not enough information to answer your question.
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: