RMSEA vs. RMSR PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
Message/Author
 Ethan Arenson posted on Wednesday, April 16, 2008 - 8:14 am
Hi, Bengt/Linda.

Can you please explain to me the difference between RMSEA and RMSR? My understanding is that Mplus reports RMSR in EFA models. I don't know how it differs from RMSEA, however.

Best,
Ethan
 Linda K. Muthen posted on Wednesday, April 16, 2008 - 8:32 am
You can find the formulas for all fit statistics in Mplus in Techncal Appendix 5 which is on the website.
 cathy labrish posted on Sunday, June 28, 2009 - 8:28 am
Hi. I'm running a very large simulation study with Mplus and wanted to know if there was anyway that I could call the Mplus program from within another program or from a DOS .BAT file so that I could automate the process and avoid having to submit each of my set of simulation commands individually. Thank-you.
 Linda K. Muthen posted on Sunday, June 28, 2009 - 9:18 am
You can run Mplus from a DOS .BAT file, for example, the .BAT file would contain the following commands:

call Mplus file1.inp
call Mplus file2.inp

where file1.inp and file2.inp are input files.
 cathy labrish posted on Monday, June 29, 2009 - 4:14 pm
I suspected as much. Thank-you very much for the speedy response!!!!! One further question: from which directory should I run the mplus directory itself?
 Linda K. Muthen posted on Monday, June 29, 2009 - 4:26 pm
It doesn't matter as long as the Mplus directory is part of your path environment variable.
 Vlad posted on Tuesday, November 03, 2009 - 12:00 pm
Hello,
I am conducting efa .All my dependent variables are categorical and I am using WLSMV estimator.My question is: should I rely on RMSEA or SRMR since RMSEA is 0,08 and SRMR is 0,03 for 3 factors?
 Linda K. Muthen posted on Tuesday, November 03, 2009 - 12:21 pm
You should also get at least chi-square and with the new rotations other fit statistics. As always in EFA, besides fit statistics, theory behind the number of factors for which the items were developed is a good guide.
 Vlad posted on Wednesday, November 04, 2009 - 3:31 am
Hi Linda,
According theory 3 factors are enough. However, Chi-Square p-value for 3factors is 0.00. Moreover, I am not sure about Chi-S. since Mplus gives some notes:* The chi-square value for MLM, MLMV, MLR, ULSMV, WLSM and WLSMV cannot be used
for chi-square difference tests. MLM, MLR and WLSM chi-square difference
testing is described in the Mplus Technical Appendices at www.statmodel.com.
See chi-square difference testing in the index of the Mplus User's Guide.

** The degrees of freedom for MLMV, ULSMV and WLSMV are estimated according to
a formula given in the Mplus Technical Appendices at www.statmodel.com.
See degrees of freedom in the index of the Mplus User's Guide.

Are p-values from Chi-S. are still reliable?

Regards,
V
 Linda K. Muthen posted on Wednesday, November 04, 2009 - 5:53 am
Yes, the p-values for all chi-square test statistics are reliable.
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: