Chi-Square Difference Testing PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Patchara Popaitoon posted on Saturday, May 05, 2012 - 12:27 pm
Dear Linda,

I use MLR for my analysis. I follow the steps provided on Mplus webpage to compute a chi-square difference test.

Is it ok to label this type of test as ..Satorra-Bentler scaled chi-squared difference test.. when I report the results?

Thanks
 Linda K. Muthen posted on Saturday, May 05, 2012 - 5:52 pm
MLR is not Satorra-Bentler. It is the Yuan-Bentler T2* test statistic.
 Patchara Popaitoon posted on Saturday, May 05, 2012 - 6:09 pm
I got it from the manual that chi-square value obtained from MLR estimates is based on Yuan-Bentler T2* test statistic. What I would like to know is whether the steps to compare nested models provided on Mplus website is based on the instruction provided by Satorra-Benter. If so, can I say that I use Satorra-Benter scaled chi-squared difference test to compare the nested models. Thanks.
 Linda K. Muthen posted on Saturday, May 05, 2012 - 9:03 pm
MLM is the Satorra-Bentler test statistic. The method proposed by Satorra and Bentler for chi-square difference testing can be used with both MLM and MLR.
 Patchara Popaitoon posted on Sunday, May 06, 2012 - 5:27 am
Thanks, Linda. This helps.
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: